Dear colleagues,
Our next BeNeRL Reinforcement Learning Seminar (Jan. 9) is coming:
Title: Diffusion Models for Decision Making
Date: January 9, 16.00-17.00 (CET)
The goal of the online BeNeRL seminar series is to invite RL researchers (mostly advanced PhD or early postgraduate) to share their work. In addition, we invite the speakers to briefly share their experience with large-scale deep RL experiments, and their style/approach
to get these to work.
We would be very glad if you forward this invitation within your group and to other colleagues that would be interested (also outside the BeNeRL region). Hope to see you on January 9!
Kind regards,
Zhao Yang & Thomas Moerland
VU Amsterdam & Leiden University
——————————————————————
Upcoming talk:
Date: January 9, 16.00-17.00 (CET)
Title: Diffusion Models for Decision Making
Abstract: Diffusion models (DMs) are powerful generative models that demonstrate promising performance across various domains. Motivated by their remarkable capability in complex distribution modeling and conditional generation, researchers have developed
a series of works applying DMs for decision-making tasks. I will introduce the various roles that DMs play in decision-making tasks from a unified perspective. I will delve into the latest advancements in aligning human feedback (AlignDiff), and increasing
decision frequency (DiffuserLite) using DMs in decision-making tasks. Finally, I will explain the framework of the unified decision diffusion models (CleanDiffuser).
Bio: Yifu Yuan is a second-year Ph.D. student at Tianjin University, under the supervision of Professor Jianye Hao. His research interests include building embodied agents for decision-making, with specific focus areas in Embodied AI and Reinforcement
Learning. He has received the National Scholarship, CIE Research Incentive Project, and was selected as a Tencent Rhino-Bird Elite Talent. He has published over 10 papers at top international AI conferences such as ICML, NeurIPS, and ICLR. He also serves as
a reviewer for multiple AI conferences and journals, such as ICML, NeurIPS, ICLR, AISTATS, and TNNLS.