Reinforcement Learning 14: Psychology
14 Psychology
14.1 Prediction and Control
14.2 Classical Conditioning
Pavlov’s dog
unconditioned stimulus, unconditioned responses
conditioned stimulus, conditioned responses
Delay Conditioning
Trace Conditioning
14.2.1 Blocking and Higher-order Conditioning
14.2.2 The Rescorla-Wagner Model
The core idea of the Rescorla-Wagner model is that an animal only learns when events violate its expectations, in other words, only when the animal is surprised.
14.2.3 The TD Model
14.2.4 TD Model Simulations
14.3 Instrumental Conditioning
Thorndike’s cat
14.4 Delayed Reinforcement
eligibility traces and value functions
14.5 Cognitive Maps
environment models
14.6 Habitual and Goal-directed Behavior
model-free and model-based