Главная
Study mode:
on
1
Intro
2
Conventional wisdom
3
What is machine learning
4
What is generalization
5
What can we take away
6
Least favorite figure
7
Inception model
8
Regularization
9
Pull Request
10
Random Features
11
Regression
12
Boosting
13
Model Size
14
Diminishing Returns
15
New Holdout Set
16
New Test Set
17
Results
18
Mechanical Turk
19
Variability
20
Imagenet Data
21
Cotton Pickers
22
Kaggle
Description:
Explore the controversial topic of training on test sets and other unconventional practices in machine learning through this thought-provoking lecture by Ben Recht from UC Berkeley. Delve into fundamental concepts of machine learning and generalization, challenging conventional wisdom in the field. Examine critical aspects such as regularization, random features, regression, boosting, and model size. Analyze the implications of diminishing returns, new holdout and test sets, and the variability in datasets like ImageNet. Gain insights into the complexities of data collection, including discussions on Mechanical Turk and Kaggle competitions. Question established norms and consider alternative approaches to improve machine learning practices and outcomes.

Training on the Test Set and Other Heresies

Simons Institute
Add to list
00:00
-00:44