0.25
0.5
0.75
1.25
1.5
1.75
2
Introduction to Machine Learning
Published on Sep 13, 201583669 Views
Related categories
Chapter list
Introduction to Machine Learning00:00
What is machine learning ?00:56
Artificial Intelliggence in the 60's03:12
Current view of ML founding disciplines03:54
What is machine-learning?04:38
The key ingredient of machinine learning is...05:36
Training data06:32
Importance of the Problem dimensions08:48
Turning data into a nice list of examples09:15
Turning an example into an input vector10:58
Dataset imagined as a point cloud in a high-dimensional vector space12:46
Ex: nearest-neighbor classifier14:04
Machine learning tasks15:02
Learning phases17:23
Ex: 1D regression18:16
Suprevised task19:49
A machine learning algorithm usually coresponds to a combination of the following 3 elements:20:56
Evaluating the quality22:21
Evaluating a predictor f(x)22:24
Standard loss-functions23:44
Surogate loss-functions25:04
Expected risk vs. Empirical risk28:14
Empirical risk minimization30:47
Evaluating the generalization error32:27
Simple train/test procedure36:04
Mode selection37:00
Ex. of parameterized function families37:34
Capacity of a learning algorithm42:16
Effective capacity, and capacity-control hyper-parameters45:59
Popular classifiers51:27
Ex. 2D classification - 101:01:02
Ex. 2D classification - 201:01:37
Ex. 2D classification - 301:01:54
Decomposing the generalization error01:02:05
Optimal capacity & the bias-variance dilemma01:07:04
Model selection how to01:09:11
Tuning the capacity01:10:56
Ex of model hyper-parameter selection01:21:57
Question01:24:11
Model selection procedure summary01:24:50
Ensemble methods01:24:59
Bagging for reducing variance on a regression problem01:26:15
How to obtain non-linear predictor with a linear predictor01:26:48
Levels of representation01:28:33
Questions?01:28:59