0.25
0.5
0.75
1.25
1.5
1.75
2
Completion of biological networks : the output kernel trees approach
Published on Apr 15, 20076490 Views
Elucidating biological networks appears nowadays as one of the most important challenge in systems biology. Due to the availability of various sources of data, machine learning has to play a major rol
Related categories
Chapter list
Completion of biological networks the output kernel00:00
Introduction00:22
Outline02:39
Outline - Supervised network inference03:05
Supervised graph inference03:07
A general solution based on considering a kernelized04:04
A kernel on graph nodes05:21
Outline - Output kernel trees05:55
Output kernel trees05:57
Back to the classic framework of supervised learning07:16
Regression trees on multiple outputs07:51
Multiple outputs regression trees09:21
Prediction with multiple outputs regression trees10:20
Output kernel trees (OK3) use the kernel trick in the output space pt 110:53
Output kernel trees (OK3) use the kernel trick in the output space pt 211:47
Algorithm devoted to the supervised network inference12:47
Regression trees in output feature space pt 114:20
Regression trees in output feature space pt 215:05
Ensemble methods16:09
Outline - Results19:04
Biological networks19:05
Comparison of different tree based methods21:52
Comparison of different sets of features23:01
Comparison with full kernel based methods23:42
Robustness26:18
Interpretability rules and clusters (an example with a protein-protein network)27:47
Interpretability feature ranking30:45
Conclusion31:15
Interpretability rules and clusters (an example with a protein-protein network) (a)33:18
Regression trees in output feature space pt 1 (a)35:15
Comparison with full kernel based methods (a)36:54