ConceptAI & ML
Importance sampling
A technique to estimate expected values under a target distribution using samples from a different proposal distribution by weighting them.
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
3Each in their own terms, from the easiest level.
Sergey Levine
David Silver
A technique to estimate the expectation of a different distribution by weighting samples by the ratio of probabilities between the target and behavior distributions.Phillip Isola
A technique to estimate an expectation by sampling from a different distribution that is more likely to produce samples that contribute significantly to the integral.
Lectures that assume it
2Connections
17The lectures that define it or assume it, and the concepts next to it.
Concepts9
Lectures8
- Model-Free Control — ε-Greedy SARSA & Off-Policy Q-Learning
- Deep Reinforcement Learning
- Exploration of fundamental concepts in deep reinforcement learning
- Deep Reinforcement Learning
- Generative Models: Representation Learning Meets Generative Modeling
- Actor-Critic Methods: Value Function Bootstrapping & Advantage Estimation