The 18th European Conference on Machine Learning (ECML) and the 11th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD)

Learning Partially Observable Markov Model from First Passage Times

author: Jerome Callut, UCL

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Learning Partially Observable Markov Models
0:01	HMM induction
0:59	Outline
1:43	FPT in models and sequences (1)
2:08	FPT in models and sequences (2)
2:18	FPT in models and sequences (3)
2:18	FPT in models and sequences (4)
2:20	FPT in models and sequences (5)
2:21	FPT in models and sequences (6)
2:22	FPT in models and sequences (7)
2:23	FPT in models and sequences (8)
2:53	FPT in models and sequences (9)
3:26	FPT in models and sequences (10)
3:57	FPT in models and sequences (11)
4:27	Partially Observable Markov Models (POMMs) (1)
4:54	Partially Observable Markov Models (POMMs) (2)
6:07	Partially Observable Markov Models (POMMs) (3)
6:51	FPT dynamics in POMMs (1)
8:58	FPT dynamics in POMMs (2)
9:48	POMM dynamics is poorly approximated by MC
11:13	POMM induction: POMMStruct (1)
12:11	POMM induction: POMMStruct (2)
12:18	Feature selection/weighting (1)
13:18	Feature selection/weighting (2)
13:37	POMM induction: POMMStruct
14:02	Parameter estimation: POMMPHit (1)
14:19	Parameter estimation: POMMPHit (2)
14:24	Parameter estimation: POMMPHit (3)
14:45	Parameter estimation: POMMPHit (4)
14:53	Parameter estimation: POMMPHit (5)
15:16	Reestimation formula
15:40	POMM induction: POMMStruct
15:59	Adding state in block (1)
16:42	Adding state in block (2)
16:46	Adding state in block (3)
17:11	Adding state in block (4)
17:34	POMM induction: POMMStruct
17:35	Experimental results
20:02	Conclusion and future work (1)
20:43	Conclusion and future work (2)

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.