Reward Prediction Error
Blake explains how a bootstrapping approach helps resolve the credit assignment problem in reinforcement learning by using immediate rewards to predict future outcomes. This leads to the concept of reward prediction error, which parallels the computations occurring in our brain's dopaminergic systems. Jon adds that both machines and humans take shortcuts to maximize rewards, highlighting the complexity of decision-making in life.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
729: Universal Principles of Intelligence (Across Humans and Machines) — with Prof. Blake Richards
Related Questions