Uploaded November 2023 | Updated November 2026, 2 days ago
Peng Ding – Associate Professor, UC Berkeley “Causal Inference in Network Experiments: Regression-Based Analysis and Design-Based Properties”
Network experiments have been widely used in investigating interference among units. Under the “approximate neighborhood interference” framework introduced by {Leung2022}, treatments assigned to individuals farther from the focal individual result in a diminished effect on the focal individual’s response, while the effect remains potentially nonzero {Leung2022} establishes the consistency and asymptotic normality of the inverse-probability weighting estimator for estimating causal effects in the presence of interference. We extend these asymptotic results to the Hajek estimator which is numerically identical to the coefficient from the weighted-least-squares fit based on the inverse probability of the exposure mapping. The numerically equivalent regression-based approach offers two notable advantages: it can provide standard error estimators through the same weighted-least-squares fit, and it allows for the integration of covariates into the analysis. Furthermore, we introduce the regerssion-based network-robust variance estimator, adopting the form of the Heteroskedasticity and Autocorrelation Consistent estimator, and analyze its asymptotic bias. Recognizing that the variance estimator can be anti-conservative, we propose an adjusted variance estimator to improve empirical coverage. Although we focus on regression-based point and variance estimators, our theory holds under the design-based framework, which assumes that the randomness comes solely from the design of network experiments and allows for arbitrary misspecification of the regression models.
Hannah Li – Assistant Professor, Columbia Business School, “When Traditional Power Analysis Falls Short – A Public Health Case Study”
A concern about service interventions, commonly found in domains like public health, education, and vocational training, is that promising interventions at the randomized controlled trial (RCT) stage may not perform well at scale. In this work we highlight and isolate the effects of one such factor: capacity constraints and the ensuing spillover effects. We consider a case study of a mobile health platform designed to improve patients’ adherence to treatment protocols. Due to capacity constraints on the system, the effect observed in an RCT may diminish when servicing a larger population. We show that this has counterintuitive implications for the typical power analysis in experiment planning and that a joint power and capacity analysis should be implemented.
Peng Ding – Associate Professor, UC Berkeley “Causal Inference in Network Experiments: Regression-Based Analysis and Design-Based Properties”
Network experiments have been widely used in investigating interference among units. Under the “approximate neighborhood interference” framework introduced by {Leung2022}, treatments assigned to individuals farther from the focal individual result in a diminished effect on the focal individual’s response, while the effect remains potentially nonzero {Leung2022} establishes the consistency and asymptotic normality of the inverse-probability weighting estimator for estimating causal effects in the presence of interference. We extend these asymptotic results to the Hajek estimator which is numerically identical to the coefficient from the weighted-least-squares fit based on the inverse probability of the exposure mapping. The numerically equivalent regression-based approach offers two notable advantages: it can provide standard error estimators through the same weighted-least-squares fit, and it allows for the integration of covariates into the analysis. Furthermore, we introduce the regerssion-based network-robust variance estimator, adopting the form of the Heteroskedasticity and Autocorrelation Consistent estimator, and analyze its asymptotic bias. Recognizing that the variance estimator can be anti-conservative, we propose an adjusted variance estimator to improve empirical coverage. Although we focus on regression-based point and variance estimators, our theory holds under the design-based framework, which assumes that the randomness comes solely from the design of network experiments and allows for arbitrary misspecification of the regression models.
Hannah Li – Assistant Professor, Columbia Business School, “When Traditional Power Analysis Falls Short – A Public Health Case Study”
A concern about service interventions, commonly found in domains like public health, education, and vocational training, is that promising interventions at the randomized controlled trial (RCT) stage may not perform well at scale. In this work we highlight and isolate the effects of one such factor: capacity constraints and the ensuing spillover effects. We consider a case study of a mobile health platform designed to improve patients’ adherence to treatment protocols. Due to capacity constraints on the system, the effect observed in an RCT may diminish when servicing a larger population. We show that this has counterintuitive implications for the typical power analysis in experiment planning and that a joint power and capacity analysis should be implemented.





![Building AI That Works for People - Dr. Rana el Kaliouby and Sinan Aral - BIG.AI@MIT 2026
In this fireside chat, Dr. Rana el Kaliouby joins MIT IDE Director Sinan Aral to discuss the future of human-AI collaboration, the gender gap in AI, and what it means to build AI that truly serves people. Kaliouby, founder of Affectiva and Managing Partner of Blue Tulip Ventures, makes the case for human-centric AI and shares her thoughts on how AI will affect the workplace and AI creativity.
Get the full BIG.AI@MIT recap here: https://bit.ly/3O6tcYn
WHAT YOU’LL LEARN
• What the future org chart looks like as AI agents join hybrid human-AI teams.
• Why AI hallucination can be a feature, not just a bug.
• Why warmth was the strongest predictor of success in fully automated AI negotiations.
CHAPTERS
0:00 Introduction: from founder to investor: Rana el Kaliouby’s journey
1:16 Human-centric AI: investing in technology that augments, not replaces
8:13 The gender gap in AI: wealth creation and who gets left out
10:27 Warmth in AI negotiations
13:02 Relationship intelligence: AI that helps you connect, not replace connection
18:55 Individual AI adoption: two siblings, two very different approaches
22:07 AI on the org chart: hybrid human-AI teams and future structures
31:32 Perception, emotion AI, and responsible design
35:06 Hallucination as creativity: rethinking a much-maligned feature
38:43 Sycophancy, manipulation, and the danger of AI that only agrees
43:31 AI companions, mental health, and the risks of over-reliance
50:12 Unlocking AI value: intangible investments CEOs need to make
1:00:33 Q&A: investing in AI
FEATURED SPEAKER
Rana el Kaliouby, Co-founder and Managing Partner, Blue Tulip Ventures
Learn more: https://ranaelkaliouby.com/
MODERATOR
Sinan Aral, Director, MIT Initiative on the Digital Economy
Learn more: https://ide.mit.edu/people/sinan-aral/
RELATED CONTENT
MIT IDE research on AI negotiations and warmth → Advancing AI Negotiations: A Large-Scale Autonomous Negotiation Competition → https://arxiv.org/abs/2503.06416
BIG.AI@MIT 2026 conference recap blog post → [ide.mit.edu URL — placeholder]
ABOUT MIT IDE
The MIT Initiative on the Digital Economy (IDE) researches how AI, data, and digital technology reshape business, work, and society. Based at MIT Sloan School of Management.
Learn more and subscribe to our newsletter: https://ide.mit.edu
#mitide #humanai #aiinvestments Building AI That Works for People - Dr. Rana el Kaliouby and Sinan Aral - BIG.AI@MIT 2026](https://i.ytimg.com/vi/P5F4gs6IWjM/mqdefault.jpg)




