Computing Up Rich Sutton Brings Reinforcements - 72nd Conversation
Dec 31, 2023
Computer scientist Rich Sutton, a quiet giant of machine learning, joins the hosts in a conversation about reinforcement learning, the potential of smarter entities, distributed computation, centralized systems, goal selection, emergence of human intelligence, centralization vs decentralization in reinforcement learning, and the nature of science.
Chapters
Transcript
Episode notes
1 2 3 4 5 6 7 8
Introduction
00:00 β’ 2min
Exploring the Potential of Smarter Entities
02:17 β’ 14min
Exploring the Concept of Distributed Computation
16:16 β’ 7min
Forces of Centralization and the Role of Reward Functions
23:12 β’ 4min
The Dynamics of Goal Selection and Sub Goals
27:08 β’ 2min
Emergence of Human Intelligence and Life from a Simplistic Goal
29:17 β’ 2min
Centralization versus Decentralization in Reinforcement Learning
31:10 β’ 21min
Exploring the Nature of Science and Excitement about a New Version of Neural Networks
52:23 β’ 2min