Name: Brianna Wu, Giant Spacekat Studios, IVECG Seminar Series Talk at UCI @UCIBrenICS
Uploaded: 2026-09-23T08:17:00-05:00
Duration: 8 min 17 s
Description: Brianna Wu, Giant Spacekat Studios, IVECG Seminar Series Talk at UCI

Brianna Wu, Giant Spacekat Studios, IVECG Seminar Series Talk at UCI

UCIBrenICS 2026-09-23 | Brianna Wu, Giant Spacekat Studios, IVECG Seminar Series Talk at UCI

Science and Storytelling: How to Make Impact with Research

UCIBrenICS 2022-12-05 | Yalda T. Uhls
Founder and Executive Director, Center for Scholars and Storytellers
UCLA

Abstract:
Streaming services have catalyzed an explosion of youth-facing content, which has massively shifted the sociocultural landscape. US youth (8-18), especially those from lower SES, spend more hours consuming media than with caregivers or in school. Media consumed during adolescence, a developmental stage of social-affective engagement and goal flexibility, may be an especially powerful socializing agent.

Researchers conduct hundreds of empirical studies containing insights into how stories could be designed for youth to minimize harms and maximize benefits. Many Hollywood creators care about prosocial learning, but without a turnkey method to integrate research findings into content, the potential impact of storytelling can go to waste. Translation work delivered in a systematic and sustained manner is critical because empirical research could be productively leveraged to shape the design, narratives, and marketing of media aimed at youth markets.

The Center for Scholars & Storytellers (CSS), based at UCLA in the psychology department, builds connections between these traditionally siloed groups. This talk will share insights into how CSS works with the entertainment and tech industry (including Disney, Mattel and YouTube) and offer case studies of impact.

Bio:
Yalda T. Uhls is an internationally recognized, award-winning research scientist, educator and author, studying how media affect young people. Her peer reviewed research has been featured in many news outlets including NPR and the NY Times. In her former career, she was a senior movie executive at MGM and Sony. Uhls is the founding director of The Center for Scholars & Storytellers, a research organization based at UCLA, which bridges the gap between social science research and media creation to support authentic and inclusive stories for youth.

Uhls is also an adjunct professor at UCLA where she does research on how media affect the social behavior of tweens and teens and teaches a class on Digital Media and Human Development and is the author of the parenting book Media Moms & Digital Dads: A Fact not Fear Approach to Parenting in the Digital Age.

Dr. Uhls knowledge of how media content is created and the science of how media affect children inform her unique perspective.

Designing Personal Informatics in the Social, Changing World

UCIBrenICS 2022-12-05 | Christina Chung
Assistant Professor, Luddy School of Informatics, Computing and Engineering
Indiana University Bloomington

Abstract:
Personal informatics refers to information individuals can collect about themselves, such as food intake, physical activity, sleep, and mood. Current personal informatics tools have been designed primarily for personal use, focusing on quantitative measurements that are easy to collect via sensors or manual input. These systems often overlook the changing nature of everyday life, the social contexts individuals live in, the variety of goals and values they have, and the constraints and preferences associated with these contexts and values. My research has examined the collaborative use of personal informatics data and co-constructed experience in various contexts. In this talk, I will share a few recent studies unpacking ways to rethink personal informatics technology that considers the changing contexts of health behavior, shifting values and priorities, as well as the social roles and relationships that often deeply intertwine with health decisions.

Bio:
Christina Chung is an Assistant Professor in Informatics and the Luddy Faculty Fellow 2020/2021 at the Indiana University Bloomington. She is also the director of the Proactive Health lab. Her research focuses on how ubiquitous computing and personal informatics data can be designed and shared to support relationships, motivate health behavior, and support collaborative care. She has published in top HCI conferences and medical journals; receiving a Best Paper Award, Honorable Mentions, and an Impact Recognition Award. Her research has been featured in mainstream media, such as CNN and Geekwire, and is supported by the National Science Foundation, IU Luddy Faculty Fellowship, and IU Precision Health Initiative.

Christina received her Ph.D. in Human Centered Design and Engineering from the University of Washington while she was a member of the Design. Use. Build (DUB) group. Previously, she was also a software engineer in IBM Research Collaboratory Taiwan conducting service innovation research in health and wellness. She holds an M.B.A and B.B.A in Information Management from the National Taiwan University.

Interaction-Centric AI: Designing Useful and Usable AI Applications

UCIBrenICS 2022-10-27 | Juho Kim
Associate Professor, KAIST; Chief Scientist, Ringle

Abstract:
AI-powered services and applications are introduced at a rapid pace and massive scale across various domains. Remarkable model performance, however, does not necessarily translate to an improved user experience. I argue that human-AI interaction should be considered a first-class object in designing AI-powered systems. In this talk, I will present a few novel interactive systems that use AI to support complex real-life tasks. I will discuss how we considered human-AI interaction in designing these systems, what tensions we encountered and how we addressed them, and how to design better AI-powered systems for real-world users. My ultimate proposal is that we need a fundamental shift to “interaction-centric AI”—an approach to systematically designing and engineering human-AI interaction that overcomes the limitations of the model- and data-centric views.

Bio:
Juho Kim [juhokim.com] is an Associate Professor in the School of Computing at KAIST, affiliate faculty in the Kim Jaechul Graduate School of AI at KAIST, and a director of KIXLAB (the KAIST Interaction Lab) [kixlab.org]. His research in human-computer interaction and human-AI interaction focuses on building interactive and intelligent systems that support interaction at scale, with the goal of improving the ways people learn, collaborate, discuss, make decisions, and take action online. He earned his Ph.D. from MIT in 2015, M.S. from Stanford University in 2010, and B.S. from Seoul National University in 2008. In 2015-2016, he was a Visiting Assistant Professor and a Brown Fellow at Stanford University. He is a recipient of KAIST’s Songam Distinguished Research Award, Grand Prize in Creative Teaching, and Excellence in Teaching Award, as well as 14 paper awards from ACM CHI, ACM CSCW, ACM Learning at Scale, ACM IUI, ACM DIS, and AAAI HCOMP. He is currently spending his sabbatical year at Ringle Inc., a startup building an online language tutoring platform, to transfer his research on automatically analyzing and diagnosing learners’ English proficiency into a real product.

Synthesis and Verification of Distributed Systems

UCIBrenICS 2022-10-12 | Mohsen Lesani
UCR

October 7, 2022
11:00am - 12:00pm

Title:
Synthesis and Verification of Distributed Systems

Abstract:
Distributed systems are the backbone of modern computing. Yet, building distributed systems with reliability and security guarantees has proven to be complicated, and remains elusive. The complication is not only faced by experts that design and implement distributed systems but is also exposed to client programmers that develop distributed applications. This talk presents an overview of our cross-stack research on mechanized verification for distributed middleware, and automatic coordination synthesis for client applications. We will then present our recent project that, in the face of attacks, assures end-to-end policies for the three aspects of trustworthiness: confidentiality, integrity and availability. Inter-organizational systems where subsystems with partial trust need to cooperate are common in healthcare, finance and military. We present the Hamraz synthesis tool that given a class and the specification of its end-to-end policies as types, applies type inference to automatically place and replicate the fields and partitioned methods of the class on Byzantine quorum systems, and synthesizes trustworthy-by-construction distributed systems. The type system provably guarantees that well-typed methods enjoy noninterference for the three properties, and that their types quantify their resilience to Byzantine attacks. The experiments show the resiliency of the resulting systems; they can gracefully tolerate attacks that are as strong as the specified policies.

Speaker Bio:
Mohsen Lesani is an associate professor at the Computer Science and Engineering department of the University of California, Riverside. He spent his postdoc at MIT and obtained his PhD from UCLA. His research interests are reliability and security of software systems especially concurrent and distributed systems. He received the NSF CAREER award in 2020 and the DARPA YFA in 2022. His research has been recognized as SIGPLAN Research Highlight in 2019 and received the distinguished paper award at OOPSLA '18.

State of the Informatics Department

UCIBrenICS 2022-10-04 | The UCI Department of Informatics is proud to present Melissa Mazmanian,
Professor, Chancellor’s Fellow and Chair, Department of Informatics at
UC Irvine.

2022 Master of Computer Science Capstone Event Highlights

UCIBrenICS 2022-07-27 | ...

2022 Donald Bren School of ICS Commencement Ceremony

UCIBrenICS 2022-06-17 | Sunday, June 12 • 5:00 pm • Bren Events Center

Congratulations, Class of 2022! You did it!

Opening Remarks: 14:27:21

National Anthem: 15:43:00
Emma Ginzel
M.F.A student, Claire Trevor School of the Arts

Dean's Remarks: 17:46:10
Marios Papaefthymiou
Dean, Donald Bren School of ICS

Student Speaker: 21:30:00
Kazeem Salaam
Baccalaureate Candidate, Donald Bren School of ICS

Featured Speaker: 28:42:00
Smita Bakshi, Ph.D. ’96
Senior VP, Wiley Technology & Engineering Careers (TEC) and co-founder of zyBooks

The Metaverses New Diversity Industrial Complex

UCIBrenICS 2022-06-15 | The UCI Department of Informatics is proud to present Lisa Nakamura, Gwendolyn Calvert Baker Collegiate Professor of American Culture, University of Michigan. This is a CREATE Distinguished Talk. The UCI campus community is invited to join us for the live talk and discussion in DBH 6011. Off-campus community is welcome to join via Zoom Meeting ID: 984 6164 8864.

Abstract: This paper analyzes the technosocial discourse surrounding the metaverse as a post-COVID platform designed for both intimacy and safety, immediacy and distance, racial and gender empathy and deniability. Companies like Oculus and Mursion create and exploit women of color avatars for use in racial sensitivity training modules and as central images in advertising campaigns to build a new diversity industrial complex without the need to hire or fairly compensate a diverse workforce.

Bio: Lisa Nakamura is the Gwendolyn Calvert Baker Collegiate Professor of American Culture at the University of Michigan and a Primary Investigator for the DISCO (Digital Inquiry, Speculation, Collaboration, and Optimism) Network, disconetwork.org, a collective of critical researchers working on race, gender, disability, and digital technologies. She is the author of several books on race, gender, and the Internet, most recently Racist Zoombombing (Routledge, 2021, co-authored with Hanah Stiverson and Kyle Lindsey) and Technoprecarious (Goldsmiths/MIT, 2020, as Precarity Lab.

2022 Engineering and ICS Hall of Fame - Full Presentation Video

UCIBrenICS 2022-06-07 | Boats bobbing atop soft ocean waves on a warm spring evening set the perfect Southern California backdrop to celebrate exceptional Anteater engineering and information and computer sciences alumni – in person for the first time since the COVID-19 pandemic hit.

More than 225 alumni, faculty and community members of the UC Irvine Samueli School of Engineering and the Donald Bren School of Information and Computer Sciences gathered to induct six alumni at the seventh annual Hall of Fame event at the Balboa Yacht Club in Corona Del Mar on May 13.

The three alumni from each school were selected for making a significant impact on their profession or bringing distinction to their alma mater. Fifty-eight engineering alumni and 46 ICS alumni have now been named Hall of Famers since it was established in 2015 to coincide with UC Irvine’s 50th anniversary.

ICS Inductees:
* Rohit Khare
Ph.D. 2003, M.S. 2000
* Srinivas Mantripragada
Ph.D. 2000
* Peyman Oreizy
Ph.D. 1999, M.S. 1995, B.S. 1993

Engineering Inductees:
* John Olivier
B.S. 1985 – Civil Engineering
* Cecilia Richards
Ph.D. 1990 – Mechanical Engineering
* Elizabeth San Miguel
B.S. 2002 – Computer Engineering

2022 Engineering and ICS Hall of Fame - Highlights Video

AI/ML Seminar Series: Bobak Pezeshki (6/06/2022)

UCIBrenICS 2022-06-06 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Bobak Pezeshki
PhD Student, Department of Computer Science
University of California, Irvine

AND/OR Branch-and-Bound for Computational Protein Design Optimizing K*

Computational protein design (CPD) is the task of creating new proteins to fulfill a desired function. In this talk, I will share work recently accepted at UAI 2022 based on a new formulation of CPD as a graphical model designed for optimizing subunit binding affinity. These new methods showed promising results when compared with state-of-the-art algorithm BBK* that is part of a long-time developed software package dedicated to CPD. In the talk, I will first describe CPD in general and for optimizing a quantity called K* (which approximates binding affinity). I will relate this to the well known task of MMAP for which many powerful algorithms have been recently developed and from which our methods are inspired. Next I will give a preview of the promising results of our new framework. I will then go on to describe the framework, presenting the formulation of the problem as a graphical model for K* optimization and introducing a weighted mini-bucket heuristic for bounding K* and guiding search. Finally, I will share our algorithm AOBB-K* and modifications that can enhance it, describing some of the empirical benefits and limitations of our scheme. To conclude, I will outline some future directions for advancing the use of this framework.

Bio: Bobak Pezeshki is a fifth year PhD student of Computer Science at the University of California, Irvine, under advisement of Professor Rina Dechter. His research focus is in automated reasoning over graphical models with focus in Abstraction Sampling and applying automated reasoning over graphical models to computational protein design. He completed his undergraduate studies at UC Berkeley majoring in Molecular and Cell Biology (with an emphasis in Biochemistry) and Integrative Biology. Before pursuing his PhD at UCI, he was involved in protein biochemistry research at the Stroud Lab, UCSF, and at Novartis Vaccines and Diagnostics.

Design as Democratic Inquiry: A Conversation with Carl DiSalvo

UCIBrenICS 2022-05-20 | Carl DiSalvo (Georgia Tech) in conversation with Paul Dourish (UC Irvine, CREATE), Leah Horgan (Northeastern University) and Lilly Irani (UC San Diego) about DiSalvo’s new book “Design as Democratic Inquiry: Putting Experimental Civics into Practice” (MIT Press, 2022).

Bio:
Carl DiSalvo is a designer, writer, researcher, and educator. He is an Associate Professor at the Georgia Institute of Technology, with appointments in the School of Interactive Computing and the School of Literature, Media, and Communication, and direct the Experimental Civics Studio. He earned a Ph.D. in Design from Carnegie Mellon University (2006). From 2006 – 2007 he was a post-doctoral fellow at Carnegie Mellon University with joint appointments in the Studio for Creative Inquiry and the Center for the Arts in Society.

He publishes regularly in design, science and technology studies, and human-computer interaction journals and conference proceedings. His first book, Adversarial Design, is part of the Design Thinking, Design Theory series at MIT Press; his second, Design as Democratic Inquiry: Putting Experimental Civics Into Action, was published this year by MIT Press. He is also a co-editor of the MIT Press journal Design Issues. His experimental design work has been exhibited and supported by the ZKM (Center for Art & Media, Karlsruhe), Grey Area Foundation for the Arts (San Francisco), Times Square Arts Alliance, Science Gallery Dublin, and the Walker Arts Center (Minneapolis).

Recorded on Thursday, May 19, 2022.

AI/ML Seminar Series: Robin Jia (5/16/2022)

UCIBrenICS 2022-05-16 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Robin Jia
Assistant Professor of Computer Science
University of Southern California

Out-of-Distribution Evaluation: The How, the Which, and the “What?!”

Natural language processing (NLP) models have achieved impressive accuracies on in-distribution benchmarks, but they are unreliable in out-of-distribution (OOD) settings. In this talk, I will give an exclusive preview of my group’s ongoing work on evaluating and improving model performance in OOD settings. First, I will propose likelihood splits, a general-purpose way to create challenging non-i.i.d. benchmarks by measuring generalization to the tail of the data distribution, as identified by a language model. Second, I will describe the advantages of neurosymbolic approaches over end-to-end pretrained models for OOD generalization in visual question answering; these results highlight the importance of measuring OOD generalization when comparing modeling approaches. Finally, I will show how synthesized examples can improve open-set recognition, the task of abstaining on OOD examples that come from classes never seen at training time.

Bio: Robin Jia is an Assistant Professor of Computer Science at the University of Southern California. He received his Ph.D. in Computer Science from Stanford University, where he was advised by Percy Liang. He has also spent time as a visiting researcher at Facebook AI Research, working with Luke Zettlemoyer and Douwe Kiela. He is interested broadly in natural language processing and machine learning, with a particular focus on building NLP systems that are robust to distribution shift. Robin’s work has received best paper awards at ACL and EMNLP.

Interrogating Maker Culture and STEAM Education

UCIBrenICS 2022-05-16 | The UCI Department of Informatics is proud to present Nettrice Gaskins, Assistant Director, Lesley University’s STEAM Learning Lab.

Abstract:
Achievement and interest gaps remain in STEM fields for students from underrepresented backgrounds. Even when students are academically successful, many still see STEM as disconnected from their daily lives and pursuits. Youth from underrepresented groups often have greater successes in out-of-school settings than in school due in part to how these settings value the end products, which are more consistent with learning in everyday life, such as identity development and vernacular practices – outcomes that align with DIY maker culture and STEAM. The book “Techno-Vernacular Creativity and Innovation” examines these practices and shows how they apply to formal and informal learning. Author Dr. Nettrice Gaskins will share some of her findings.

Bio:

Dr. Gaskins teaches, writes, "fabs”, and makes art using algorithms and machine learning. She has taught multimedia, visual art, and computer science with high school students. She earned a BFA in Computer Graphics with Honors from Pratt Institute in 1992 and an MFA in Art and Technology from the School of the Art Institute of Chicago in 1994. She received a doctorate in Digital Media from Georgia Tech in 2014. Currently, Dr. Gaskins is a 2021 Ford Global Fellow and the assistant director of the Lesley STEAM Learning Lab at Lesley University. She is an advisory board member for the School of Literature, Media, and Communication at Georgia Tech. Her first full-length book, Techno-Vernacular Creativity and Innovation is available through The MIT Press. Gaskins' AI-generated artworks can be viewed in journals, magazines, museums, and on the Web. Her series of 'featured futurist' portraits are on view at the Smithsonian Arts and Industries Building through early July 2022.

Media Ruins: Infrastructural Restitution and Building Futures in Post-Conflict Cambodia

UCIBrenICS 2022-05-16 | The UCI Department of Informatics is proud to present Maggie Jack, Postdoctoral Scholar, Syracuse University/UC Irvine.

Abstract: This talk describes the ways that Cambodian new media creators commemorate lost artists and an imagined better way of life through finding, repairing, and disseminating historical film, photography and cinema artifacts from before the Khmer Rouge period, often using digital tools. Reconstructing such media artifacts through a process of infrastructural restitution is a mode of healing from decades of national conflict and a form of subtle political action in an increasingly authoritarian Phnom Penh. Building on theory at the intersection of infrastructure studies (Star and Ruhleder, 1996; Larkin, 2013) and media’s relationship to memory (Gordon, 2008; Larkin, 2008; Richards, 1994), the concept of infrastructural restitution allows us to (re)integrate the importance of memory, the affective, and the spiritual into scholarship of infrastructure. This case gives new insight into the tension in transnational technology use between creative appropriation and the problematic political economy of mainstream platforms. The empirical sections of this talk are based on my historical and ethnographic research in Phnom Penh beginning in January 2014, including 20 months of full-time research from June 2017-January 2019.

Bio: I am a postdoctoral scholar on the NSF-funded project “Creating Work/Life” with a team spanning Syracuse University (PI: Ingrid Erickson) and University of California, Irvine (PI: Melissa Mazmanian). I am a research affiliate at the Digital Life Initiative at Cornell Tech in New York City and an adjunct professor at NYU Tandon, teaching “Transnational Technology” in the spring of 2022. I hold a PhD in Information Science (2020) from Cornell University, where I had a minor PhD concentration in Anthropology and was an active member of the Southeast Asia Program. I use my past professional experiences in the technology industry in Silicon Valley and the international development sector and my academic background in the History of Science (BA Harvard University; MPhil University of Cambridge) to approach questions of contemporary computing with both scholarly and practical lenses. My writing is published in the Proceedings of the SIGCHI Conference on Human Factors in Computing (CHI), Interactions Magazine, The Information Society, Global Perspectives, Computer Supported Cooperative Work (CSCW), and elsewhere. My book-in-progress Media Ruins is under contract in the Labor and Technology series at the MIT Press (Katie Helke, editor; Winifred Poster, series editor).

AI/ML Seminar Series: Ties van Rozendaal (5/09/2022)

UCIBrenICS 2022-05-09 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Ties van Rozendaal
Senior Machine Learning Researcher
Qualcomm AI Research

Instance-adaptive data compression: Improving Neural Codecs by Training on the Test Set

Neural data compression has been shown to outperform classical methods in terms of rate-distortion performance, with results still improving rapidly. These models are fitted to a training dataset and cannot be expected to optimally compress test data in general due to limitations on model capacity, distribution shifts, and imperfect optimization. If the test-time data distribution is known and has relatively low entropy, the model can easily be finetuned or adapted to this distribution. Instance-adaptive methods take this approach to the extreme, adapting the model to a single test instance, and signaling the updated model along in the bitstream. In this talk, we will show the potential of different types of instance-adaptive methods and discuss the tradeoffs that these methods pose.

Bio: Ties is a senior machine learning researcher at Qualcomm AI Research. He obtained his masters’s degree at the University of Amsterdam with a thesis on personalizing automatic speech recognition systems using unsupervised methods. At Qualcomm AI research he has been working on neural compression, with a focus on using generative models to compress image and video data. His research includes work on semantic compression and constrained optimization as well as instance-adaptive and neural-implicit compression.

AI/ML Seminar Series: Maurizio Filippone and Ba-Hien Tran (5/02/2022)

UCIBrenICS 2022-05-02 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Maurizio Filippone, Associate Professor, EURECOM
Ba-Hien Tran, PhD Student, EURECOM

Functional Priors for Bayesian Deep Learning

The Bayesian treatment of neural networks dictates that a prior distribution is specified over their weight and bias parameters. This poses a challenge because modern neural networks are characterized by a huge number of parameters and non-linearities. The choice of these priors has an unpredictable effect on the distribution of the functional output which could represent a hugely limiting aspect of Bayesian deep learning models. Differently, Gaussian processes offer a rigorous non-parametric framework to define prior distributions over the space of functions. In this talk, we aim to introduce a novel and robust framework to impose such functional priors on modern neural networks for supervised learning tasks through minimizing the Wasserstein distance between samples of stochastic processes. In addition, we extend this framework to carry out model selection for Bayesian autoencoders for unsupervised learning tasks. We provide extensive experimental evidence that coupling these priors with scalable Markov chain Monte Carlo sampling offers systematically large performance improvements over alternative choices of priors and state-of-the-art approximate Bayesian deep learning approaches.

Bio: Maurizio Filippone received a Master’s degree in Physics and a Ph.D. in Computer Science from the University of Genova, Italy, in 2004 and 2008, respectively. In 2007, he was a Research Scholar with George Mason University, Fairfax, VA. From 2008 to 2011, he was a Research Associate with the University of Sheffield, U.K. (2008-2009), with the University of Glasgow, U.K. (2010), and with University College London, U.K (2011). From 2011 to 2015 he was a Lecturer at the University of Glasgow, U.K, and he is currently AXA Chair of Computational Statistics and Associate Professor at EURECOM, Sophia Antipolis, France. His current research interests include the development of tractable and scalable Bayesian inference techniques for Gaussian processes and Deep/Conv Nets with applications in life and environmental sciences.
Bio: Ba-Hien Tran is currently a PhD student within the Data Science department of EURECOM, under the supervision of Professor Maurizio Filippone. His research focuses on Accelerating Inference for Deep Probabilistic Modeling. In 2016, he received a Bachelor of Science degree with honors in Computer Science from Vietnam National University, HCMC. His thesis investigated Deep Learning approaches for data-driven image captioning. In 2020, he received a Master of Science in Engineering degree in Data Science from Télécom Paris. His thesis focused on Bayesian Inference for Deep Neural Networks.

Combating Flaky Tests

UCIBrenICS 2022-04-21 | The UCI Department of Informatics is proud to present:
Darko Marinov
Professor of Computer Science
University of Illinois at Urbana-Champaign

Pivoting a Business Multiple Times and Staying a Fan of Your Idea

UCIBrenICS 2022-04-20 | ICS Alumni Chapter's Lunch & Learn series presents:

"Pivoting a Business Multiple Times and Staying a Fan of Your Idea"

Featuring:
Vijesh Mehta '01 (CEO, EZ Texting) and
Punit Shah '03 (CMO, EZ Texting)

Recorded on Friday, April 15, 2022.

Vijesh and Punit have spent more than 15 years building a VC backed company from startup to where it is today. Along the way, they were forced to reinvent themselves and the company. Hear how pivoting the business and themselves were key lessons and inflection points in business. We'll be talking with them and discussing topics from startup ideas, scaling, acquisitions, and business models.

Speaker Bios:
Vijesh Mehta is the CEO, Chairman & Co-founder of EZ Texting. He co-founded the company in 2005, which started as a self-service marketing solution used by businesses to better and more effectively communicate with their customers. Over the past 15 years, Vijesh has held multiple leadership roles within the EZ Texting. He has been instrumental in the growth of the company, including securing two rounds of investor-led financing, driving the business to profitability, and promoting the cultural values enabling the company to receive several distinguished “Best Places to Work” awards in Los Angeles and Austin.

Punit Shah is an expert at creating, transforming, and maximizing data-driven, low friction businesses. Shah has had repeat success in creating industry leading marketing efficiency through online channels and developing high ROI, easy to use, online products to meet the needs of wide underserved markets. Shah’s leadership approach drives growth through a harmonious combination of scalable data analysis & reporting, crisp goals, powerful vision, deep empathy, and strong culture. Shah is an active advisor to companies that engage his interests and expertise. He also spends as much of his time as possible mentoring and guiding his contacts on how to achieve the next phase of professional and personal growth.

Geometric Aggregation of the Social Welfare in Resource Allocation

UCIBrenICS 2022-04-12 | The UCI department of Computer Science is proud to present Yinyu Ye,
Stanford University.

ABSTRACT:
We present efficient algorithms for resource allocation to optimize the geometrically aggregated social-welfare objective function. Unlike the usual (weighted) arithmetic average objective, the (weighted) geometric average provides some ideal features such as strong concavity, social-fairness, and decentralization property, as demonstrated in the classic Fisher market equilibrium. In this talk, we show:

1) Complexity of computing an optimal solution for the weighted geometric average objective is in the same class of linear programming that uses the arithmetic welfare average.
2) It can be computed in a distributed fashion by using the primal-dual and/or ADMM methods while preserving individual utility privacy.
3) It can be implemented in the online setting with a sublinear regret, such that exhibited in online linear programming.

The major takeaway from this talk: it is desirable and doable for optimization/decision models that uses geometrically aggregated social wel:fare objectives.

SPEAKER BIO:
Yinyu Ye is currently the K.T. Li Chair Professor of Engineering at Department of Management Science and Engineering and Institute of Computational and Mathematical Engineering, Stanford University. He received the B.S. degree in System Engineering from the Huazhong University of Science and Technology, China, and the M.S. and Ph.D. degrees in Engineering-Economic Systems and Operations Research from Stanford University. His current research interests include Continuous and Discrete Optimization, Data Science and Application, Algorithm Design and Analysis, Computational Game/Market Equilibrium, Metric Distance Geometry, Dynamic Resource Allocation, and Stochastic and Robust Decision Making, etc. He is an INFORMS (The Institute for Operations Research and The Management Science) Fellow since 2012, and has received several academic awards including: the 2009 John von Neumann Theory Prize for fundamental sustained contributions to theory in Operations Research and the Management Sciences, the 2015 SPS Signal Processing Magazine Best Paper Award, the winner of the 2014 SIAM Optimization Prize awarded (every three years), the inaugural 2012 ISMP Tseng Lectureship Prize for outstanding contribution to continuous optimization (every three years), the inaugural 2006 Farkas Prize on Optimization, the 2009 IBM Faculty Award, etc.. He has supervised numerous doctoral students at Stanford who received various prizes such as INFORMS Nicholson Prize, Student Paper Competition, the INFORMS Computing Society Prize, the INFORMS Optimization Prize for Young Researchers.

8 Common Characteristics ofPeople Who Are Successful at Entrepreneurship & Life

UCIBrenICS 2022-04-12 | The UCI ICS Alumni Chapter is proud to present:

Featuring Jeff Greenberg, M.S. '84,
Managing Director, Tech Coast Works

The Entrepreneurial Learning Initiative has identified 8 characteristics that are common in successful entrepreneurs. It turns out that these characteristics are also indicative of success in life in general. Jeff will describe these 8 characteristics and then lead a discussion about the importance of these characteristics and an exploration of which of these characteristics exist in each of us.

Speaker Bio: Jeff has been fortunate enough to have experienced success in three phases of his career, initially as a corporate executive, then as an entrepreneur and currently as a consultant and professor of innovation and entrepreneurship. Jeff is one of the most successful entrepreneurs in the history of UCI, having paid royalties of about $1M from companies he founded based on UCI licensed technologies.
As a consultant, he helps tech entrepreneurs around the world navigate the path from invention to commercial success. He teaches on various innovation and entrepreneurship topics at UCI, Cal State Fullerton and Irvine Valley College. Jeff holds a BS in computer science from Rutgers, an MS in computer science from UCI and an MBA from Pepperdine.

Intersecting Social Media and Games Research to Center Inclusion in Media Design

UCIBrenICS 2022-04-12 | The UCI Department of Informatics is proud to present Jennifer Pierre,
User Experience Researcher, Google.

ABSTRACT: This talk will walk through findings from Dr. Pierre's dissertation work on youth social media use among minoritized youth, and her more recent work developing a system for evaluating diversity, equity, and inclusion (DEI) in games. These findings will be used to highlight how these two areas of research intersected and influenced each other to provide the foundation for thoroughly centering the needs and values of minoritized communities in the design of social and digital media. Defining inclusion in games is a complex and challenging space, and learning from previous work across media studies as well as acknowledging the intersections between games and other forms of media were both integral pieces of the success of the recent inclusion in games project. Applying theory from these intersecting research areas enables researchers to advocate for the needs of minoritized communities in daily work across social media and game design.

BIOS: Jennifer Pierre, Ph.D., is a user experience and human-computer interaction researcher with expertise in social media, games, critical data studies, and social informatics. Her research explores how people, especially underrepresented and minoritized groups, use various forms of media and data to form and maintain communities. She received her doctorate from the Department of Information Studies at UCLA, and holds an MLIS from the same department. Jennifer is currently working as a User Experience Researcher at YouTube. Her work can be found in several top journals and conferences, including CHI, HICSS, PACMHCI CSCW, and Big Data & Society, and has been recognized by ACM SIGCHI, the Ford Foundation, and the Bouchet Honor Society. In addition to and intersected with her research and industry work, Jennifer has fueled her passion for inclusion in STEM, higher education, and tech as a member and leader of several diversity, equity, and inclusion initiatives.

A Call to Action!

UCIBrenICS 2022-04-12 | The UCI department of Informatics is proud to present Ron Baecker,
Emeritus Professor of Computer Science and Bell Chair in HCI
University of Toronto.

ABSTRACT: Our world has been animated and enriched by digital technologies used for creativity, collaboration, learning, health, politics, and commerce. Yet there is much that is troubling.
We depend upon software that nobody truly understands and that is vulnerable to hackers and cyberterrorism. Privacy has been overrun by governments and surveillance capitalism. Our children are addicted to their devices; we have become workaholics. Jobs and livelihoods are being demolished without adequate social safety nets. A few digital technology leviathans threaten to control not only their domains, but all commerce. There is huge hype associated with modern AI, and many risks to society stemming from its premature use before it is ready.
Happily, we are not helpless victims of forces totally outside our control. We can raise our voices as citizens; we can enact laws and be proactive as a society. My talk will mention steps of both kinds, then focus on what we as digital technology and information society professionals can and must do.

BIO: Ron Baecker is Emeritus Professor of Computer Science and Bell Chair in Human-Computer Interaction at the University of Toronto. His B.Sc., M.Sc., and Ph.D. (1969) are from MIT. He was the co-founder of UofT’s Dynamic Graphics Project (DGP), and the founder of its Knowledge Media Design Institute (KMDI) and its Technologies for Aging Gracefully lab (TAGlab).
He has been named one of the 60 Pioneers of Computer Graphics by ACM SIGGRAPH. He is an ACM Fellow, an ACM Distinguished Speaker. and a Canadian Digital Media Pioneer. He has been elected to the CHI Academy by ACM SIGCHI, and was given a 2020 CHI Social Impact Award.
His most recent books are:
Baecker, R.M. (2019). Computers and Society: Modern Perspectives, Oxford University Press.
Baecker, R.M., Feldman, G., Langer, J., and Stein, J. (2020). The COVID-19 Solutions Guide: Health, Wealth, Technology, and the Human Spirit.
Baecker, R.M. (2021). Digital Dreams Have Become Nightmares: What We Must Do.
He is also the founding Editor of the Synthesis Lectures on Technology and Health (Springer Nature, Publisher), and the organizer of the virtual community and resource hub, computers-society.org.

Navigating Trauma in Role-Playing Narrative

UCIBrenICS 2022-03-11 | The UCI GDIM Speaker Series is proud to present Shelly Jones, SUNY Delhi Professor.

Angels of Efficiency. Big Data, Visual Information Systems, and Consulting Before the Computer

UCIBrenICS 2022-03-07 | The UCI Department of Informatics is proud to preset Florian Hoof , Research Associate, Institute of Advanced Study on Media Cultures of Computer Simulation, Leuphana University of Lüneburg.

Abstract:
My talk looks at big data practices at the turn of the 20th century and how they emerged and solidified within the context of business consulting. While histories of big data and information technology are typically tied to the advent of cybernetics and computer technology, I deploy a different, sociomaterial genealogy of big data practices. I explore the ways in which the latest media technology, and avant-garde aesthetics, economic pressures, and holistic philosophy together constituted the form of consulting dominant today, and which consequences arise from this. It shows how visual charting, film, simulation devices, and calculation devices were used by consultants such as Lillian and Frank Gilbreth and Henry Gantt in business organizations. It describes the installation of planning and charting rooms, centralized spaces in which business data were collated and visualized. With this, different scenarios could be devised, graphically compared, and interpolated into the future. This form of visual management led to new business forecasting services and enabled a fast reaction to production disruptions, which needed to be facilitated in planning processes and accounting. Early big data practices created new ways to automatically unify, standardize, select and process information. Procedures and forms of data arise that share some similarities with concepts of computational data.

book link

Bio:
Florian Hoof is a research associate at Institute of Advanced Study on Media Cultures of Computer Simulation, Leuphana University Lueneburg and an associated lecturer at Goethe-University Frankfurt. His research interests include media history, non-theatrical film, organization theory and digital cultures. He is the author of Angels of Efficiency. A Media History of Consulting (Oxford University Press 2020) and of Slippery Media (Palgrave 2022). Furthermore, he is coeditor of the forthcoming book Films that Work Harder. The Global Circulation of Industrial Cinema (Amsterdam University Press 2022). Recently published articles include: Culture, Technology, and Process in ‘Media Theories’: Toward a Shift in the Understanding of Media in Organizational Research. In: Organization 26(5) 636–654, 2019.

Watching D&D Online

UCIBrenICS 2022-03-02 | The GDIM Speaker Series is proud to present Adrianna Burton, Doctoral student and streaming role-play expert at UC Irvine.

Understanding Indie Role-Playing Games

UCIBrenICS 2022-03-01 | The GDIM Speaker Series is proud to present: Evan Torner, Professor, University of Cincinnati.

AI/ML Seminar Series: Sunipa Dev (2/28/2022)

UCIBrenICS 2022-02-28 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Sunipa Dev
Research Scientist
Ethical AI Team, Google AI

Towards Inclusive and Socially Aware Language Technologies

Large language models are commonly used in different paradigms of natural language processing and machine learning, and are known for their efficiency as well as their overall lack of interpretability. Their data driven approach for emulating human language often results in human biases being encoded and even amplified, potentially leading to cyclic propagation of representational and allocational harm. We discuss in this talk some aspects of detecting, evaluating, and mitigating biases and associated harms in a holistic, inclusive, and culturally-aware manner. In particular, we discuss the disparate impact on society of common language tools that are not inclusive of all gender identities.

Bio: Sunipa Dev is a Research Scientist on the Ethical AI team at Google AI. Previously, she was an NSF Computing Innovation Fellow at UCLA, before which she completed her PhD at the University of Utah. Her ongoing research focuses on various facets of fairness and interpretability in NLP, including robust measurements of bias, cross-cultural understanding of concepts in NLP, and inclusive language representations.

CS Distinguished Seminar: Judea Pearl (2/25/2022)

UCIBrenICS 2022-02-25 | UCI CS Seminar Series
https://www.cs.uci.edu/events/seminar-series/

The Science of Causal and Effect: From Deep Learning to Deep Understanding
Judea Pearl
UC Los Angeles

Abstract:
We will define "deep understanding" as the capacity to answer questions at all three levels of the reasoning hierarchy: predictions, actions, and imagination. Accordingly I will describe a language, calculus and algorithms that facilitate all three modes of reasoning. The talk will then summarize several reasoning tasks that have benefitted from this calculus, including attribution, mediation, data-fusion and missing-data. I will conclude with future applications, which include: automated scientific explorations, personalized decision making and social intelligence

Bio:
Judea Pearl is Chancellor professor of computer science and statistics at UCLA, where he directs the Cognitive Systems Laboratory and conducts research in artificial intelligence, human cognition, and philosophy of science.

He has authored three fundamental books, Heuristics (1983), Probabilistic Reasoning (1988) and Causality (2000, 2009) which won of the London School of Economics Lakatos Award in 2002. More recently, he co-authored Causal Inference in Statistics (2016, with M. Glymour and N. Jewell) and "The Book of Why" (2018, with Dana Mackenzie) which brings causal analysis to a general audience.

Pearl is a member of the National Academy of Sciences the National Academy of Engineering, a Fellow of the Cognitive Science Society, the Royal Statistical Society, and the Association for the Advancement of Artificial Intelligence. In 2012, he won the Technion's Harvey Prize and the ACM Alan Turing Award "for fundamental contribution to artificial intelligence through the development of a a calculus for probabilistic and causal reasoning." In 2022 he won the BBVA Frontiers of Knowledge Award for “laying the foundations of modern artificial intelligence, so computer systems can process uncertainty and relate causes to effects.”

Website: http://bayes.cs.ucla.edu/jp_home.html

Founding Wizards of the Coast

UCIBrenICS 2022-02-23 | The GDIM Speaker Series is proud to present Peter Adkinson, Founder of Wizards of the Coast and owner of Gen Con.

Designing Opportunities for AI and UX in Healthcare

UCIBrenICS 2022-02-22 | The UCI Department of Informatics is proud to present:
Jofish Kaye & Erika Poole
Senior Director, Interaction Design & Artificial Intelligence &
Director, User Research
Anthem Health Platforms

Abstract:
What if doctors could pick the right medicines for you, not based on a limited clinical trial of 500 people, or the few hundred people they’d personally seen in their career, but instead on the medical records of millions of people with the same diseases in the real world? Machine learning has made such advances possible, but that work raises very real problems when end users, like doctors and nurses, interact with recommendations based on inference from large datasets, rather than from clinical trials. In this talk, Jofish and Erika talk about their work at Anthem Health Platforms at the intersection of AI, UR, IxD, and healthcare: about the processes to make that work happen, some guidelines for building such interfaces, and the implications of this work for HCI and Informatics.

Bio:
Jofish Kaye, Ph.D, is Senior Director of Interaction Design & Artificial Intelligence at Anthem Health Platforms. He runs research teams to produce thoughtful and ethical HCI and AI products. He has served on the ACM Diversity & Inclusion Council, chaired CHI 2016, and has a long running interest in improving diversity, inclusion, and accessibility. He has a Ph.D in Information Science from Cornell and MS and BS degrees from MIT. 

Erika Poole is Director of User Research at Anthem Health Platforms. She's passionate about removing complexity & confusion in healthcare. Erika has a PhD in Human-Centered Computing and MS in Computer Science from Georgia Tech, and BS in Computer science from Purdue University.

Developing Reinforcement Learning Agents that Learn Many Subtasks

UCIBrenICS 2022-02-18 | The UCI Department of Computer Science is proud to present Martha White, Associate Professor of Computing Science, University of Alberta.

Title:
Developing Reinforcement Learning Agents that Learn Many Subtasks

Abstract:
Learning agents operating in complex environments must accumulate knowledge about the environment to continually improve. This knowledge can take the form of a dynamics model, option policies that achieve certain subgoals and long-term predictions in the form of general value functions. All of these are subtasks that the agent can learn about in parallel, to improve performance on the primary task: accumulating reward. When we commit to the perspective that our reinforcement learning agents need to discover, learn and use many subtasks, new algorithmic considerations arise. The agent needs to answer: how can I direct data gathering (exploration) to learn these subtasks efficiently? How can I learn these subtasks in parallel, from a single stream of experience, and maintain stability under these off-policy (counterfactual) updates? In this talk, I will motivate the need to develop such agents, as well as insights into how to efficiently learn these subtasks using directed exploration and off-policy algorithms.

Speaker Bio:
Martha White is an Associate Professor of Computing Science at the University of Alberta and a PI of Amii--the Alberta Machine Intelligence Institute--which is one of the top machine learning centres in the world. She holds a Canada CIFAR AI Chair and received IEEE's "AIs 10 to Watch: The Future of AI" award in 2020. She has authored more than 50 papers in top journals and conferences. Martha is an associate editor for TPAMI, and has served as co-program chair for ICLR and area chair for many conferences in AI and ML, including ICML, NeurIPS, AAAI and IJCAI. Her research focus is on developing algorithms for agents continually learning on streams of data, with an emphasis on representation learning and reinforcement learning.

Rethinking Representation in Role-Playing Games

UCIBrenICS 2022-02-16 | The GDIM Speaker Series is proud to present Edmond Chang, Professor, Ohio University.

AI/ML Seminar Series: Ruiqi Gao (2/14/2022)

UCIBrenICS 2022-02-14 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Ruiqi Gao
Research Scientist
Google Brain

Advanced training of energy-based models

Energy-based models (EBMs) are an appealing class of probabilistic models, which can be viewed as generative versions of discriminators, yet can be learned from unlabeled data. Despite a number of desirable properties, two challenges remain for training EBMs on high-dimensional datasets. First, learning EBMs by maximum likelihood requires Markov Chain Monte Carlo (MCMC) to generate samples from the model, which can be extremely expensive. Second, the energy potentials learned with non-convergent MCMC can be highly biased, making it difficult to evaluate the learned energy potentials or apply the learned models to downstream tasks.
In this talk, I will present two algorithms to tackle the challenges of training EBMs. (1) Diffusion Recovery Likelihood, where we tractably learn and sample from a sequence of EBMs trained on increasingly noisy versions of a dataset. Each EBM is trained with recovery likelihood, which maximizes the conditional probability of the data at a certain noise level given their noisy versions at a higher noise level. (2) Flow Contrastive Estimation, where we jointly estimate an EBM and a flow-based model, in which the two models are iteratively updated based on a shared adversarial value function. We demonstrate that EBMs can be trained with a small budget of MCMC or completely without MCMC. The learned energy potentials are faithful and can be applied to likelihood evaluation and downstream tasks, such as feature learning and semi-supervised learning.

Bio: Ruiqi Gao is a research scientist at Google, Brain team. Her research interests are in statistical modeling and learning, with a focus on generative models and representation learning. She received her Ph.D. degree in statistics from the University of California, Los Angeles (UCLA) in 2021 advised by Song-Chun Zhu and Ying Nian Wu. Prior to that, she received her bachelor’s degree from Peking University. Her recent research themes include scalable training algorithms of deep generative models, variational inference, and representational models with implications in neuroscience.

Learning Algorithms and Complex systems: Alloys

UCIBrenICS 2022-02-10 | The UCI Department of Computer Science is proud to present:
Petros Koumoutsakos
Herbert S. Winokur, Jr. Professor of Engineering and Applied Sciences, John A. Paulson School of Engineering and Applied Sciences
Harvard University

Recorded:
February 4, 2022
11:00am - 12:00pm

Title:
Learning Algorithms and Complex systems: Alloys

Abstract:
Over the last thirty years we have experienced more than a billion-fold increase in hardware capabilities and a dizzying pace of acquiring and transmitting massive amounts of data. Learning algorithms have been the beneficiaries of these advances and today they are increasingly embedded in technologies that touch every aspect of humanity. However along with the abundance of promise there is an ever increasing amount of hype, in particular regarding the capabilities of learning algorithms to model, predict and control complex physical systems.
In this talk I would offer a perspective on forming alloys of learning algorithms and simulations for the prediction and control of complex systems. I will present novel algorithms and a comparative study of back-propagation algorithms for the modeling of chaotic dynamical systems, a fusion of reinforcement learning and scientific computing for modeling and control of complex flow-structure interactions. I will juxtapose successes and failures and argue that the proper fusion of domain knowledge and machine learning expertise are essential to advance human knowledge.

Speaker Bio:
Petros Koumoutsakos is Herbert S. Winokur, Jr. Professor of Engineering and Applied Sciences, Faculty Director of the Institute for Applied Computational Science (IACS) and Department Chair of Applied Mathematics at Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS). He studied Naval Architecture (Diploma-NTU of Athens, M.Eng.-U. of Michigan), Aeronautics and Applied Mathematics (PhD-Caltech) and has served as the Chair of Computational Science at ETH Zurich (1997-2020). Petros is elected Fellow of the American Society of Mechanical Engineers (ASME), the American Physical Society (APS), the Society of Industrial and Applied Mathematics (SIAM). He is recipient of the Advanced Investigator Award by the European Research Council and the ACM Gordon Bell prize in Supercomputing. He is elected International Member to the US National Academy of Engineering (NAE). His research interests are on the fundamentals and applications of computing and artificial intelligence to understand, predict and optimize fluid flows in engineering, nanotechnology, and medicine.

Authoring Role-Playing Worlds

UCIBrenICS 2022-02-09 | The GDIM Speaker Series is proud to present:
Matt Forbeck, NY Times bestselling author

Designing Games that Resist (2/8)

UCIBrenICS 2022-02-09 | The GDIM Speaker Series is proud to present:
Jason Morningstar, award-winning designer of games such as Fiasco

AI/ML Seminar Series: Maja Rudolph (2/07/2022)

UCIBrenICS 2022-02-07 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Maja Rudolph
Senior Research Scientist
Bosch Center for AI

Modeling Irregular Time Series with Continuous Recurrent Units

Recurrent neural networks (RNNs) are a popular choice for modeling sequential data. Standard RNNs assume constant time-intervals between observations. However, in many datasets (e.g. medical records) observation times are irregular and can carry important information. To address this challenge, we propose continuous recurrent units (CRUs) – a neural architecture that can naturally handle irregular intervals between observations. The CRU assumes a hidden state which evolves according to a linear stochastic differential equation and is integrated into an encoder-decoder framework. The recursive computations of the CRU can be derived using the continuous-discrete Kalman filter and are in closed form. The resulting recurrent architecture has temporal continuity between hidden states and a gating mechanism that can optimally integrate noisy observations. We derive an efficient parametrization scheme for the CRU that leads to a fast implementation (f-CRU). We empirically study the CRU on a number of challenging datasets and find that it can interpolate irregular time series better than methods based on neural ordinary differential equations.

Bio: Maja Rudolph is a Senior Research Scientist at the Bosch Center for AI where she works on machine learning research questions derived from engineering problems: for example, how to model driving behavior, how to forecast the operating conditions of a device, or how to find anomalies in the sensor data of an assembly line. In 2018, Maja completed her Ph.D. in Computer Science at Columbia University, advised by David Blei. She holds a MS in Electrical Engineering from Columbia University and a BS in Mathematics from MIT.

Reinventing You for the New Year: A Journey of Growth

UCIBrenICS 2022-02-07 | The UCI ICS Alumni Chapter is proud to present Jamar T. Balacy, UX Designer, Amazon.

Join us as Jumar shares his personal journey and how he has helped himself and others re-invent themselves to ultimately find their passions. Although career transitions and life changing events can be intimidating, Jumar shares how the right mindset, preparation, and reflection are the keys to ultimately redefining yourself.

Speaker Bio:
Jumar is a UX Designer working in tech and holds a BS in ICS from UC Irvine. Currently, he works at Amazon designing internal tools for Operations and Supply Chain. He has also held a variety of different positions before going all-in with UX. His past roles include Data Analyst, Tech PM, and Front End Developer. He was also a US Navy Combat Camera Diver where he documented light search and recovery missions, underwater damage assessment, and anti-terrorism force protection searches keeping piers and harbors safe. His passions include mentoring new designers and advocating for veteran recruiting initiatives in the workplace.

AI/ML Seminar Series: Dylan Slack (1/31/2022)

UCIBrenICS 2022-01-31 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Dylan Slack
PhD Student
Department of Computer Science
University of California, Irvine

Exposing Shortcomings and Improving the Reliability of Machine Learning Explanations

For domain experts to adopt machine learning (ML) models in high-stakes settings such as health care and law, they must understand and trust model predictions. As a result, researchers have proposed numerous ways to explain the predictions of complex ML models. However, these approaches suffer from several critical drawbacks, such as vulnerability to adversarial attacks, instability, inconsistency, and lack of guidance about accuracy and correctness. For practitioners to safely use explanations in the real world, it is vital to properly characterize the limitations of current techniques and develop improved explainability methods. This talk will describe the shortcomings of explanations and introduce current research demonstrating how they are vulnerable to adversarial attacks. I will also discuss promising solutions and present recent work on explanations that leverage uncertainty estimates to overcome several critical explanation shortcomings.

Bio: Dylan Slack is a Ph.D. candidate at UC Irvine advised by Sameer Singh and Hima Lakkaraju and associated with UCI NLP, CREATE, and the HPI Research Center. His research focuses on developing techniques that help researchers and practitioners build more robust, reliable, and trustworthy machine learning models. In the past, he has held research internships at GoogleAI and Amazon AWS and was previously an undergraduate at Haverford College advised by Sorelle Friedler where he researched fairness in machine learning.

Game Design Process

UCIBrenICS 2022-01-27 | The GDIM 59 lecture course is proud to present Alex Roberts, designer of the Star Crossed RPG.

A Human-Centered Approach Towards Creating An Inclusion and Equitable Society

UCIBrenICS 2022-01-25 | The UCI Informatics Department is proud to present Earl Huff Jr. , Doctoral Candidate and Research Assistant, Clemson University.

Title:
"To Understand The Problem, You Must Understand The People: A Human-Centered Approach Towards Creating An Inclusion and Equitable Society"

Abstract:
Computer and information technology are increasingly becoming an important factor in shaping how society advances across several domains. In transportation, for example, we see advances in automated driver assistance systems and autonomous driving capabilities to reduce vehicle accidents and save lives. In education, we see in-classroom technology such as smart boards to enhance the quality of instruction and electronic tools to enable remote learning. As society increases to rely on technology, so will the increase of qualified researchers, designers, and developers. Hence, every year jobs in the computing and tech industry are among the highest growth in the U.S.

However, computing has historically limited the participation of specific populations from accessing and benefiting from opportunities in the field due to differences in privilege, access, and awareness. The resulting lack of diversity in computing affects who can develop technology and how technology is developed to be usable and accessible for all potential users. Certain technologies, such as desktop applications and touchscreen user interfaces, may be inaccessible for people with disabilities, and A.I.-based technologies may be trained on bias data due to a lack of diverse perspectives. Without diversity in computing, there is a risk that the resulting technology may be created without critical perspectives and may unintentionally provide inequitable user experiences for consumers.

This talk discusses current efforts to expand the participation of marginalized groups in computing in terms of access to education and technology. Discussion of groups includes racial and ethnic minorities, women, people with disabilities, and aging adults. Past and existing interventions helped increase students' awareness, agency, and self-efficacy in pursuing a computing career. Designing and developing technology to be inclusive and accessible for everyone is demonstrated with applications of user-centered design (UCD) to consider the perspectives and needs of future users.

Bio:
Earl Huff Jr. is a Doctoral Candidate studying Human-Centered Computing in the School of Computing at Clemson University. He is a Research Assistant in the Design and Research of In-Vehicle Experiences (DRIVE) Lab, directed by his advisor Dr. Julian Brinkley. Prior to starting the Ph.D., Earl earned his Master's and Bachelor's degree in Computer Science from Rowan University. His research is at the intersection of human-computer interaction, computing education, and broadening participation. Earl focuses on human-centered approaches in creating inclusive and equitable technology for all users and diversifying participation in computing for marginalized populations. His work has focused on racial and ethnic minorities, women, and people with disabilities. Earl's work has been published in high-quality venues such as ASSETS, the Technical Symposium for Computer Science Education, the International Conference on Software Maintenance and Evolution, and AutomotiveUI.

AI/ML Seminar Series: Ransalu Senanayake (1/24/2022)

UCIBrenICS 2022-01-24 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Ransalu Senanayake
Postdoctoral Scholar
Department of Computer Science
Stanford University

Propagating Uncertainty from Modeling into Decision-Making for Trustworthy Autonomy

Autonomous agents such as self-driving cars have already gained the capability to perform individual tasks such as object detection and lane following, especially in simple, static environments. While advancing robots towards full autonomy, it is important to minimize deleterious effects on humans and infrastructure to ensure the trustworthiness of such systems. However, for robots to safely operate in the real world, it is vital for them to quantify the multimodal aleatoric and epistemic uncertainty around them and use that uncertainty for decision-making. In this talk, I will talk about how we can leverage tools from approximate Bayesian inference, kernel methods, and deep neural networks to develop interpretable autonomous systems for high-stakes applications.

Bio: Ransalu Senanayake is a postdoctoral scholar in the Statistical Machine Learning Group at the Department of Computer Science, Stanford University. He focuses on making downstream applications of machine learning trustworthy by quantifying uncertainty and explaining the decisions of such systems. Currently, he works with Prof. Emily Fox and Prof. Carlos Guestrin. He also worked on decision-making under uncertainty with Prof. Mykel Kochenderfer. Prior to joining Stanford, Ransalu obtained a PhD in Computer Science from the University of Sydney, Australia, and an MPhil in Industrial Engineering and Decision Analytics from the Hong Kong University of Science and Technology, Hong Kong.

RPGs and Brand Capitalism

UCIBrenICS 2022-01-24 | The GDIM 53: Roleplaying and Improvisational Play speaker series is proud to present:

Title:
"RPGs and Brand Capitalism"

Speaker:
Professor Megan Condis
Texas Tech University

Recorded:
Thursday, 1/20

Black Aesthetics: Interventions in Digital Media

UCIBrenICS 2022-01-18 | The UCI Informatics Department is proud to present the Winter Quarter 2022 Informatics Seminar Series.

Title:
“Black Aesthetics: Interventions in Digital Media”

Speaker:
A.M. Darke
Assistant Professor of Performance, Play and Design, Digital Arts and New Media, and Critical Race and Ethnic Studies
UC Santa Cruz

Abstract:
Across digital media, Black people are portrayed in ways that are derogatory, inaccurate, stereotypical, demeaning, and otherwise harmful–if we are depicted at all. The representation of afro-textured hair is noticeably limited, with options ranging from comically large afros, unstyled “dread” locs, and misshapen cornrows. Black character models for games and 3D media are often one-dimensional racial caricatures; the savage, the criminal, the mammy, the athlete, the fetish, the muscle, the braggart.

Through projects like ‘Ye or Nay? and the Open Source Afro Hair Library, artist A.M. Darke discusses her critical approaches to constructing Black identity in virtual space.

Speaker Bio:
A.M. Darke is an artist and game maker designing radical tools for social intervention. Still in the class war. Now in the pandemic. He’s in the combination class war and pandemic. Assistant Professor of Performance, Play & Design, Digital Arts and New Media, and Critical Race and Ethnic Studies, at UC Santa Cruz, Darke also directs The Other Lab, an interdisciplinary, intersectional feminist research space for experimental games and new media. Darke’s recent work includes ‘Ye or Nay?, a Kanye West-themed game about Black culture, and the Open Source Afro Hair Library, a 3D model database for Black hair styles and textures. Darke holds a B.A. in Design (’13) and an M.F.A. in Media Arts (’15), both from UCLA. He has completed residencies with Laboratory, NYU Game Center, the Open Data Institute, and the Frank-Ratchye STUDIO for Creative Inquiry at Carnegie Mellon. Additionally, his work has been shown internationally and featured in a variety of publications, including Kill Screen, Vice, and NPR.

AI/ML Seminar Series: Roy Fox (1/10/2022)

UCIBrenICS 2022-01-10 | UCI AI/ML Seminar Series
https://cml.ics.uci.edu/aiml/

Roy Fox
Assistant Professor
Department of Computer Science
University of California, Irvine

Curiously effective ensemble and double-oracle reinforcement-learning methods

Ensemble methods for reinforcement learning have gained attention in recent years, due to their ability to represent model uncertainty and use it to guide exploration and to reduce value estimation bias. We present MeanQ, a very simple ensemble method with improved performance, and show how it reduces estimation variance enough to operate without a stabilizing target network. Curiously, MeanQ is theoretically *almost* equivalent to a non-ensemble state-of-the-art method that it significantly outperforms, raising questions about the interaction between uncertainty estimation, representation, and resampling.
In adversarial environments, where a second agent attempts to minimize the first’s rewards, double-oracle (DO) methods grow a population of policies for both agents by iteratively adding the best response to the current population. DO algorithms are guaranteed to converge when they exhaust all policies, but are only effective when they find a small population sufficient to induce a good agent. We present XDO, a DO algorithm that exploits the game’s sequential structure to exponentially reduce the worst-case population size. Curiously, the small population size that XDO needs to find good agents more than compensates for its increased difficulty to iterate with a given population size.

Bio: Roy Fox is an Assistant Professor and director of the Intelligent Dynamics Lab at the Department of Computer Science at UCI. He was previously a postdoc in UC Berkeley’s BAIR, RISELab, and AUTOLAB, where he developed algorithms and systems that interact with humans to learn structured control policies for robotics and program synthesis. His research interests include theory and applications of reinforcement learning, algorithmic game theory, information theory, and robotics. His current research focuses on structure, exploration, and optimization in deep reinforcement learning and imitation learning of virtual and physical agents and multi-agent systems.

UCIBrenICS Live Stream

UCIBrenICS 2022-01-09 | ...

The Promise of Access: Technology, Inequality, and the Political Economy of Hope - Daniel Greene

UCIBrenICS 2021-12-07 | The UCI Department of Informatics is proud to present:

Daniel Greene
Assistant Professor, Information Studies
University of Maryland

“The Promise of Access: Technology, Inequality, and the Political Economy of Hope”

Abstract:
Why do we keep trying to solve poverty with technology? What makes us feel that we need to learn to code—or else? This common sense has ruled our economic imaginary for at least 30 years. Those who cannot log on or train up are condemned to the margins of the information economy, and contained by the carceral state.

In The Promise of Access, Daniel Greene argues that the problem of poverty became a problem of technology in order to manage the contradictions of a changing economy. We cannot debunk or banish the idea—what Greene calls the access doctrine—that the problem of poverty can be solved with the right tools and the right skills because the idea helps those public institutions that face poverty to save themselves. Technological solutions help public institutions simplify their complex missions and win legitimacy and funding, but at the cost of alienating the populations they serve.

Blending political-economic theory with years of ethnographic fieldwork, Greene explores how this plays out in Washington, DC, examining organizational change in technology startups, public libraries, and charter schools. Tracing the changes to the spirit and structure of these public institutions changes reveals a fight to define the good life under contemporary capitalism--and the alliances that could win that fight.

Bio:
Daniel Greene is an Assistant Professor of Information Studies at the University of Maryland. His ethnographic, historical, and theoretical research explores how the future of work is built and who is included in that future. He published his first book, The Promise of Access: Technology, Inequality, and the Political Economy of Hope, with MIT Press in 2021. His research has also appeared in such venues as Research in the Sociology of Work, New Media & Society, and the International Journal of Communication. Daniel lives online at dmgreene.net.

The Naturalness and Artifice of Code: Exploiting the Bimodality - Prem Devanbu, UC Davis

UCIBrenICS 2021-12-07 | The UCI Department of Informatics is proud to present:

Prem Devanbu
Professor, Department of Computer Science
UC Davis

“The Naturalness and Artifice of Code: Exploiting the Bimodality”

Abstract:
While natural languages are rich in vocabulary and grammatical flexibility, most human are mundane and repetitive. This repetitiveness in natural language has led to great advances in statistical NLP methods.

In our lab, we discovered (a decade ago) that, despite the considerable power and flexibility of programming languages, large software corpora are actually even more repetitive than NL Corpora. We went on to show that this “naturalness” of code could be captured in statistical models, and exploited within software tools. This line of work has been turbo-charged by the tremendous capacity and design flexibility of deep learning models. Numerous other creative and interesting applications of naturalness have ensued, from colleagues around the world, and several industrial applications have emerged. Recently, we have been studying the consequences and opportunities arising from the observation that Software is bimodal: it's written not only to be run on machines, but also read by humans; this makes software amenable to both algorithmic analysis, and statistical prediction. Bimodality allows new ways of training machine learning models, new ways of designing analysis algorithms, and new ways to understand the practice of programming. In this talk, I will begin with a backgrounder on "Naturalness" studies, and the promise of bimodality.

Bio:
Prem Devanbu earned his B.Tech from IIT Madras, and a Ph.D from Rutgers University. After many years developing software for Bell Laboratories and offshots in New Jersey, he joined UC Davis where he conducts teaching & research in software engineering. He has won several awards for his work, including several best paper awards, distinguished paper awards, most influential paper awards, and test-of-time awards. Three of his papers were invited to appear in CACM Research Highlights. He served as PC Chair of ESEC/FSE 2006 and ICSE 2010, and also as GC of MSR 2014 and ESEC/FSE 2020. He has served on the Editorial boards of ACM TOSEM, IEEE ToSE, the JSME, and the EMSE Journal; he serves currently on the CACM Editorial Board. He has been an ACM Fellow since 2018, and won the ACM SIGSOFT Outstanding Research Award in 2021. He even has his own web page.

Detecting Ballot Manipulations With a Transparent Voting Machine - Juan E. Gilbert

UCIBrenICS 2021-12-07 | The UCI Department of Informatics is proud to present:

Juan E. Gilbert
Andrew Banks Family Preeminence Endowed Professor and Chair, Computer and Information Science and Engineering Department
University of Florida

"Detecting Ballot Manipulations With a Transparent Voting Machine"

Abstract:
Touch-screen ballot-marking devices (BMDs) produce paper ballots that are counted by optical-scan voting machines and can be recounted by hand. If the BMD is hacked or misprogrammed so that it prints a different candidate selection than the voter indicated on the touchscreen, the voter is supposed to notice this. Recent studies suggested that only a small fraction of voters read their paper ballot carefully enough to catch errors. Furthermore, after the 2020 Presidential Election allegations of ballot manipulations have spread via social media questioning the integrity of our elections. In the talk, I will present a new design for ballot marking devices, the transparent voting machine. User studies on the prototype show a dramatically higher rate for voter detection of errors using the transparent voting machine interface.

Bio:
Dr. Juan E. Gilbert is the Andrew Banks Family Preeminence Endowed Professor and Chair of the Computer & Information Science & Engineering Department at the University of Florida where he leads the Human Experience Research Lab. He is a Fellow of the Association of Computing Machinery (ACM), a Fellow of the American Association of the Advancement of Science (AAAS), and a Fellow of the National Academy of Inventors (NAI). Dr. Gilbert is the inventor of Prime III, an open source, secure and accessible voting technology that has been used in numerous organization elections and recently in statewide elections in New Hampshire and Butler County, Ohio. Prime III is the only open-source voting system to be used in state, local and federal elections in the U.S.A. Dr. Gilbert was a member of the National Academies Committee on the Future of Voting: Accessible, Reliable, Verifiable Technology that produced the report titled, "Securing the Vote: Protecting American Democracy".