[07] John Schulman - Optimizing Expectations: From Deep RL to Stochastic Computation Graphs
The Thesis Review
00:00
Journey Through AI: From Neuroscience to Reinforcement Learning
This chapter navigates the intricate challenges of framing the PhD journey within the context of reinforcement learning, emphasizing the significance of short-term rewards for achieving long-term goals. The speaker shares their personal evolution from an initial fascination with artificial intelligence to a deeper focus on reinforcement learning, influenced by their academic experiences at Berkeley. It also highlights the transformative impact of deep learning advances on their research, particularly in robotics and the adaptation of reinforcement learning algorithms.
Transcript
Play full episode
