ConceptEconomics
POMDP
A partially observable Markov decision process is a generalization of an MDP where the agent cannot directly observe the underlying state.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
Lecturers
2Each in their own terms, from the easiest level.
Hamsa Bastani
A generalization of a Markov decision process that models an agent decision-making process where the system dynamics are partially observable.
Lectures that assume it
5Connections
14The lectures that define it or assume it, and the concepts next to it.
Concepts7
- Markov decision process
- Probability distribution
- Belief-space hierarchical planning
- History states
- Information-gathering actions
- Model-based offline RL for multi-turn dialogue
Show 1 moreShow fewer
Lectures7
- LLM-Guided Reinforcement Learning for Mastery Learning: Large-Scale Field Evidence
- Output feedback
- Challenges & Open Problems in Deep Reinforcement Learning
- The Role of Rationality in Modern Robotics
- RL with Sequence Models & Language Models, Part 1 (CS 285, Lecture 21)
- RL with Sequence Models & Language Models, Part 3