ActInf GuestStream 108.1 ~ Next-Generation AI for Emergent Semantic Communications (Christo Thomas) @ActiveInference
ActInf GuestStream 108.1 ~ Next-Generation AI for Emergent Semantic Communications (Christo Thomas)  @ActiveInference
Uploaded May 2025 | Updated November 2026, 16 hours ago
Christo Kurisummoottil Thomas
"Next-Generation Artificial Intelligence for Emergent Semantic Communications"

Creating artificial intelligence (AI)-native next-generation wireless networks (e.g., 6G) faces technical hurdles due to data-driven, training-intensive AI limitations, such as black-box models, limited reasoning and adaptability, data dependency, and energy inefficiency. To overcome these challenges, we propose that next-generation wireless systems must embrace advanced AI models that mimic human-like reasoning and planning capabilities. Along this direction, in this talk, we present a forward-looking framework grounded in causal machine learning, which is founded on the principles of causal (cause and effect relations) discovery, representation learning, and inference. Such causal inference frameworks are poised to facilitate dynamic adaptability, resilience, cognition, and time-sensitive operations in future-generation networks, thanks to explainability, generalizability and sustainability induced by causality. Further, we illustrate the potential of causal reasoning AI frameworks through its application in an emerging paradigm known as semantic communications. In particular, we introduce a pioneering approach called "emergent semantic communication," by leveraging signaling games, and causal machine learning to craft a nuanced semantic language between the communicating nodes that mimics human-like communication. This language aims to optimize semantic representation, that reduces transmission by incorporating reasoning components at both ends of the communication process. Moreover, we present an example illustrating how semantic abstractions facilitate planning-defined as a sequence of network and control actions-to enhance the quality of experience for a remote autonomous agent controlled by a base station. Finally, we will also look at future directions, including designing emergent language for multiuser semantic communications using active inference framework.

Active Inference Institute information:
Website: https://www.activeinference.institute/
Activities: https://activities.activeinference.institute/
Discord: https://discord.activeinference.institute/
Donate: http://donate.activeinference.institute/
YouTube: youtube.com/c/ActiveInference
X: https://x.com/InferenceActive
Active Inference Livestreams: https://video.activeinference.institute/
ActInf GuestStream 108.1 ~ Next-Generation AI for Emergent Semantic Communications (Christo Thomas)ActInf GuestStream 109.1 ~ Hemming: Free Energy, Attentional Narrowing, and Cognitive RigidityTao Gong & Hangjun Chu: “What bounds the surprisal? Zones of Bounded Surprisal”Fundamentals of Active Inference (Chapter 2, Session 10) May 15, 2026Alexander Sabine: “FEP and Active Inference through the lens of Winnicott’s Potential Space”6th Applied Active Inference Symposium (Day 2, Nov 13, 2026) ~ LIVEActInf GuestStream 132.1: Root Frequency Theory: A Multiscale Framework for Measuring Coherence...ActInf ModelStream 007.3: pymdp 1.0.0 (Conor Heins et al.)ActInf GuestStream 015.4 Bobby Azarian: A Strange Loop Solution to the Hard Problem of ConsciousnessFundamentals of Active Inference (Chapter 1 and Appendix A, Session 5) April 30, 2026Fundamentals of Active Inference (Chapter 1 and Appendix A, Session 4) April 28, 2026ActInf Textbook Group ~ Cohort 8 ~ Session 5 (Chapter 3) ~ 3/20/2025
Active Inference Institute |

ActInf GuestStream 108.1 ~ Next-Generation AI for Emergent Semantic Communications (Christo Thomas)

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER