The 7th International Symposium on Intelligent Data Analysis

A Partial Correlation-Based Algorithm for Causal Structure Discovery with Continuous Variables

author: Jean-Philippe Pellet, IBM Zurich Research Lab

Description

We present an algorithm for causal structure discovery suited in the presence of continuous variables. We test a version based on partial correlation that is able to recover the structure of a recursive linear equations model and compare it to the well-known PC algorithm on large networks. PC is generally outperformed in run time and number of structural errors.

You might be experiencing some problems with Your Video player.

*Slides*
0:00	A Partial Correlation-Based Algorithm to Causal Structure Learning
0:07	Outline pt 1
0:35	Outline pt 2
0:37	Causal Inference and Causal Graphs pt 1
1:17	Causal Inference and Causal Graphs pt 2
1:32	Usefulness of a Causal Model pt 1
2:00	Usefulness of a Causal Model pt 2
2:12	Usefulness of a Causal Model pt 3
2:29	Illustration: Simpson’s Paradox pt 1
2:58	Illustration: Simpson’s Paradox pt 2
3:43	Properties of the Dataset pt 1
4:09	Properties of the Dataset pt 2
4:41	Properties of the Dataset pt 3
4:46	Outline pt 3
4:51	Structure Learning Algorithms pt 1
5:38	Structure Learning Algorithms pt 2
6:13	Distinguishing Causes from Effects pt 1
7:14	Distinguishing Causes from Effects pt 2
7:38	Typical Structure Learning Algorithm
8:49	Continuity pt 1
9:41	Continuity pt 2
9:57	Outline pt 4
10:02	Total Conditioning (TC) Algorithm pt 1
11:32	Total Conditioning (TC) Algorithm pt 2
11:35	Total Conditioning (TC) Algorithm pt 3
12:07	The Alarm Network
12:48	Results: Alarm, Errors Against Sample Size
13:59	Results: Alarm, Run Time Against Sample Size
14:37	Conclusion pt 1
15:17	Conclusion pt 2
15:42	Conclusion pt 3
16:06	Thank You for Your Attention!
17:57	The Alarm Network (a)
19:31	Total Conditioning (TC) Algorithm pt 2 (a)

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.