Скачать или смотреть Foundations of Reinforcement Learning | Deep Learning Meetup with Professor Francisco Melo

Timestamps:
0:00 - Introduction
3:10 - Meetup Outline
5:25 - RL intuition
9:25 - What is RL?/Key concepts
15:24 - Policies/How to evaluate an RL agent?
20:15 - Q-Values
24:34 - How to solve RL problems?
26:10 - Estimation of Q-Values
27:35 - Monte-Carlo RL
29:50 - TD-Learning (SARSA)
32:30 - Q-learning
35:35 - Deep RL
40:15 - Deep Q-learning "Deadly Triad"
43:47 - Policy Learning
48:00 - REINFORCE
53:02 - RL applications & examples
59:03 - RL challenges
1:02:25 - Wrap-up
1:03:40 - Q&A
1:19:20 - Final remarks


You've probably heard how Reinforcement Learning systems have beaten even the world's best players at games like Go and StarCraft II. Besides mastering mind-boggling games, these systems are also used in robotics. But how do they learn so effectively by interacting with their environment? How can Reinforcement Learning be combined with Deep Learning? Luckily, professor Francisco Melo will tell us all about it in our next meetup!

Francisco Melo is a Senior Researcher at GAIPS, INESC-ID and an Associate Professor at the Computer Science Department of Instituto Superior Técnico (IST), where he teaches courses on Decision making and AI (  / francisco-melo-aba83b8  .

As always, expect up to 1h of talk, with around 10 minutes in the end for the Q&A (be sure to leave all the questions you might have during the talk on the YouTube chat!).

Links:
The presentation content is available on our GitHub page: https://github.com/DeepLearningLisbon​
Application form to apply to speak at one of our meetups: https://forms.gle/rwqE9aUY1gxZDtvq7​
Feedback form to help us improve on future events: https://forms.gle/68Udh75QTE4kMoHz6