TalkRL: The Reinforcement Learning Podcast Robert Lange
44 snips
Dec 20, 2021 Robert Tjarko Lange, a PhD student at TU Berlin, discusses topics like meta reinforcement learning, hard-coded behaviors in animals, lottery ticket hypothesis and pruning masks in deep RL, semantic RL with action grammars, advances in meta RL, the need for scientific governance, and exploring the role of parameterization in RL.
Chapters
Transcript
Episode notes
1 2 3 4 5 6 7
Introduction
00:00 • 3min
Meta Reinforcement Learning and Hard-Coded Behaviors
02:49 • 10min
Lottery Ticket Hypothesis and Pruning Masks
12:56 • 15min
Semantic Reinforcement Learning with Action Grammars
28:04 • 15min
Advances in Meta Reinforcement Learning
43:13 • 24min
The Need for Scientific Governance and Educating Future Leaders
01:06:46 • 2min
Exploring the Role of Parameterization in Reinforcement Learning
01:08:19 • 3min