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

Efficient Weight Learning for Markov Logic Networks

author: Daniel Lowd, University of Washington

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Efficient Weight Learning for Markov Logic Networks
0:20	Outline
0:34	Markov Logic Networks
1:52	Example: WebKB (1)
2:57	Example: WebKB (2)
3:22	Overview
4:22	Sneak preview
5:16	Outline - Algorithms
5:18	Gradient descent
5:28	Gradient descent in MLNs
7:03	Per-weight learning rates
8:28	Ill-Conditioning
9:14	The Hessian matrix
10:05	Newton’s method
10:42	Diagonalized Newton’s method
11:10	Conjugate gradient
11:54	Scaled conjugate gradient
12:11	Step sizes and trust regions
14:30	Preconditioning
15:13	Outline - Experiments
15:16	Experiments: Algorithms
15:47	Experiments: Datasets
16:57	Experiments: Method
17:20	Results: Cora AUC (1)
17:27	Results: Cora AUC (2)
17:43	Results: Cora AUC (3)
17:54	Results: Cora AUC (4)
18:19	Results: Cora CLL (1)
18:23	Results: Cora CLL (2)
18:35	Results: Cora CLL (3)
18:39	Results: Cora CLL (4)
18:49	Results: WebKB AUC (1)
19:13	Results: WebKB AUC (2)
19:26	Results: WebKB AUC (3)
19:37	Results: WebKB CLL
19:46	Conclusion

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.