Unsupervised Zero-Shot Reinforcement Learning via Functional Reward Encodings.
Kevin FransSeohong ParkPieter AbbeelSergey LevinePublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- supervised learning
- unsupervised learning
- function approximation
- model free
- reinforcement learning algorithms
- semi supervised
- learning algorithm
- state space
- eligibility traces
- reinforcement learning methods
- data driven
- temporal difference
- reward function
- markov decision processes
- dynamic programming
- learning problems
- optimal control
- multi agent
- average reward
- dynamical systems
- total reward
- robotic control
- optimal policy
- learning process
- object recognition
- machine learning
- decision problems
- non binary
- neural network
- orders of magnitude