COMPSCI 188 - 2018-09-27 - Reinforcement Learning Part 2/2

Рет қаралды 24,379

5 жыл бұрын

COMPSCI 188, LEC 001 - Fall 2018
COMPSCI 188, LEC 001 - Pieter Abbeel, Daniel Klein
Copyright @2018 UC Regents; all rights reserved
"Slides (from 2018): inst.eecs.berkeley.edu/~cs188/fa18
Latest website: inst.eecs.berkeley.edu/~cs188
More resources: ai.berkeley.edu
00:00 Setup [no content]
02:00 Announcements [outdated]
04:20 Review: RL
08:21 Review: Q-Learning
16:57 Exploration vs. Exploitation, Epsilon Greedy
22:19 Exploration Functions
30:59 Regret
32:55 Generalizing Across States
38:07 Feature-Based Representations
41:31 Break [no content]
45:58 Approximate Q-Learning
51:44 Example: Approximate Q-Learning
55:33 Q-Learning and Least Squares
1:05:46 Policy Search
1:08:30 Example: Helicopter
1:13:43 More RL Applications
1:20:08 End [no content]"

Пікірлер: 15

@deeplearningpartnership 5 жыл бұрын

Thank you so much for these.

@SaisankarGochhayat 5 жыл бұрын

Actual Lecture starts - 4:22

@jayanthramachandra7859 4 жыл бұрын

Break Ends 45:58

@shamsansari 10 ай бұрын

58:00 I thought you square it so larger errors have a bigger proportion than smaller errors

@quiteSimple24 5 жыл бұрын

2:02 start

@oxrsc 7 ай бұрын

31:15 what happened in washington 2018 september?

@cicartaya 3 ай бұрын

Follow me here. Let's assume by "Washington", he meant "Washington, D.C." This would imply that statement has something to do with politics. We know that UC Berkeley is in California, a largely liberal state. Recall that Trump, a highly polarizing person, was president between 2017 through 2021. Given that this video is from 2018, it seems reasonable to assume that he's talking about Trump being in Washington... But I could be wrong lol