Lecture 19 - Optimization and Learning for Robot Control - Dynamic Programming and Monte Carlo
Автор: Andrea Del Prete
Загружено: 2025-11-07
Просмотров: 58
This lecture starts with a recap on Markov Decision Processes. Then we move on to Dynamic Programming in the infinite horizon setting, discussing Iterative Policy Evaluation, Policy Iteration, Modified Policy Iteration, and finally Value Iteration.
In the last part of the lecture we discuss how Monte Carlo can be used for estimating the Value function of a fixed policy.
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке: