Chapter
Deep Reinforcement Learning and Neural Networks
26:15 - 40:16 (14:01)
Deep reinforcement learning is a family of solution methods that leverage the representation power of neural networks to learn functions for different components of the agent, including the value function, the policy, and the model of the environment. Despite the non-linear and bumpy nature of neural networks, deep learning has proven to be a universal toolkit for representing any function and making progress in learning.
Clips
The guest speaker took a year off to study intelligence with a renowned game designer, Rich Satin, and shares their experience of discovering which path to take.
26:15 - 28:02 (01:47)
Summary
The guest speaker took a year off to study intelligence with a renowned game designer, Rich Satin, and shares their experience of discovering which path to take.
ChapterDeep Reinforcement Learning and Neural Networks
Episode#86 – David Silver: AlphaGo, AlphaZero, and Deep Reinforcement Learning
PodcastLex Fridman Podcast
The challenge in solving the problem of artificial intelligence is to find a single clear problem definition that can be formalized, for which a solution would mean AI has been cracked.
28:02 - 34:39 (06:36)
Summary
The challenge in solving the problem of artificial intelligence is to find a single clear problem definition that can be formalized, for which a solution would mean AI has been cracked. Reinforcement learning can be used to decompose this problem into smaller pieces that work together to solve the larger problem of AI.
ChapterDeep Reinforcement Learning and Neural Networks
Episode#86 – David Silver: AlphaGo, AlphaZero, and Deep Reinforcement Learning
PodcastLex Fridman Podcast
Deep Reinforcement Learning is an approach in solving reinforcement learning problems that utilizes neural networks to represent any component of the agent like the policy, model, or value function that can learn and represent knowledge better and better by utilizing more resources, computation, data and experience from the environment.
34:39 - 40:16 (05:37)
Summary
Deep Reinforcement Learning is an approach in solving reinforcement learning problems that utilizes neural networks to represent any component of the agent like the policy, model, or value function that can learn and represent knowledge better and better by utilizing more resources, computation, data and experience from the environment.