AI & MLRecommendedAIML-RL-014
Temporal-Difference Learning
Bootstrapping from the next estimate: TD(0) and SARSA learn online, step by step.
- Difficulty
- 3 / 5
- Importance
- 4 / 5
- Study time
- 1–2 h
- Kind of work
- Mixed
Why it matters
The learner can compute a TD error and update and explain how bootstrapping trades Monte Carlo variance for bias.