From kernels to causal inference thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

From kernels to causal inference

Published on Jan 25, 20127682 Views

Kernel methods in machine learning have expanded from tricks to construct nonlinear algorithms to general tools to assay higher order statistics and properties of distributions. They find applications

Related categories

Chapter list

From Kernels to Causal Inference00:00
Dependence vs. Causation00:55
Statistical Implications of Causality02:05
Functional Causal Model03:17
Functional Model, ctd.04:38
Causal Inference from Observational Data05:59
Restricting the Functional Model07:54
Causal Inference with Additive Noise, 2-Variable Case09:47
Identifiability Result11:08
Causal Inference Method12:37
Experiments - 113:08
Experiments - 213:30
Experiments - 313:47
Independence-based Regression14:42
Kernel Independence Testing15:36
The Kernel Trick17:54
Detection of Confounders18:04
Inferring deterministic causal relations19:30
Causal independence implies anticausal dependence - 121:28
Causal independence implies anticausal dependence - 222:46
80 Cause-Effect Pairs24:17
80 Cause-Effect Pairs − Examples - 124:35
80 Cause-Effect Pairs − Examples - 224:56
Causal Learning and Anticausal Learning26:54
Covariate Shift and Semi-Supervised Learning29:20
Lens error correction - 133:32
Lens error correction - 234:08
Lens error correction - 334:47
Inverting a nontrivial convolution model - 135:09
Inverting a nontrivial convolution model - 235:15
Blurred image35:23
Our approach35:26
Raw Sequence35:30
Shift-Invariant Kernel Mean Maps35:49
Causal Inference for Individual Objects37:01
Causal Markov Conditions - 138:00
Causal Markov Conditions - 238:27
Kolmogorov complexity38:49
Conditional Kolmogorov complexity39:07
Algorithmic mutual information39:20
Conditional algorithmic mutual information39:40
Algorithmic mutual information: example39:54
Postulate: Local Algorithmic Markov Condition39:54
Equivalence of Algorithmic Markov Conditions40:10
Algorithmic model of causality40:41
Applications41:41
Acknowledgements41:48