Online Reinforcement Learning and Sequential Forecasting and Partial Feedback

author: Peter Auer, University of Leoben

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Pump-Priming Projects “Online Performance of Reinforcement Learning” & “Sequential Forecasting and Partial Feedback”
0:11	Online performance of reinforcement learning with internal reward functions - 1
0:12	Online performance of reinforcement learning with internal reward functions - 2
0:13	Online performance of reinforcement learning with internal reward functions - 3
0:14	Online performance of reinforcement learning with internal reward functions - 4
0:16	Online performance of reinforcement learning with internal reward functions - 5
0:17	Online performance of reinforcement learning with internal reward functions - 6
0:40	Sequential Forecasting and Partial Feedback: Applications to Machine Learning
1:42	Activities (partial list)
2:30	Scientific outcome (partial list)
2:46	ML algorithms for parameter optimization: UCT
4:02	Example for parameter optimization with UCT - 1
4:12	Example for parameter optimization with UCT - 2
4:34	Example for parameter optimization with UCT - 3
5:11	Example for parameter optimization with UCT - 4
5:41	Example for parameter optimization with UCT - 5
6:09	Example for parameter optimization with UCT - 6
6:21	Application of UCT parameter optimization - 1
6:45	Application of UCT parameter optimization - 2
7:04	Apprenticeship learning using inverse reinforcement learning and gradient methods - 1
7:25	Apprenticeship learning using inverse reinforcement learning and gradient methods - 2
7:35	Apprenticeship learning using inverse reinforcement learning and gradient methods - 3
7:51	Apprenticeship learning using inverse reinforcement learning and gradient methods - 4
8:07	Solving the IRL task
9:33	IRL – Experiments
11:02	Reinforcement Learning
12:32	Undiscounted online regret - 1
13:04	Undiscounted online regret - 2
13:54	Bounds on the regret - 1
15:01	Bounds on the regret - 2
15:55	Relation to other work: PAC-like bounds - 1
16:49	Relation to other work: PAC-like bounds - 2
17:12	Relation to other work: PAC-like bounds - 3
17:46	Relation to other work: PAC-like bounds - 4
18:43	Relation to other work: log T bounds - 1
18:59	Relation to other work: log T bounds - 2
19:14	Relation to other work: log T bounds - 3
19:40	Relation to other work: log T bounds - 4
20:48	The UCRL algorithm: Upper Confidence Reinforcement Learning
22:18	Details of UCRL: The bias - 1
22:37	Details of UCRL: The bias - 2
22:50	Details of UCRL: The bias - 3
23:09	Details of UCRL: The bias - 4
23:27	Details of UCRL: The bias - 5
23:39	Details of UCRL: Analysis - 1
23:42	Details of UCRL: Analysis - 2
23:45	Details of UCRL: Analysis - 3
23:48	Details of UCRL: Analysis - 4
23:52	Details of UCRL: Analysis - 5
24:00	Future research - 1
24:05	Future research - 2
24:32	Future research - 3
24:42	Future research - 4

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.