Machine Learning Summer School on Theory and Practice of Computational Learning

Theory, Methods and Applications of Active Learning

author: Rui Castro, Department of Electrical Engineering, Columbia University
author: Robert Nowak, Department of Electrical and Computer Engineering, University of Wisconsin - Madison

Description

Traditional approaches to machine learning and statistical inference are passive, in the sense that all data are collected prior to analysis in a non-adaptive fashion. One can envision, however more active strategies in which information gleaned from previously collected data is used to guide the selection of new data. This talk discusses the emerging theory of such "active learning" methods. I will show that feedback between data analysis and data collection can be crucial for effective learning and inference. The talk will describe two active learning problems. First, I will consider binary-valued prediction (classification) problems, for which the prediction errors of passive learning methods can be exponentially larger than those of active learning. Second, I will discuss the role of active learning in the recovery of sparse vectors in noise. I will show that certain weak, sparse patterns are imperceptible from passive measurements, but can be recovered perfectly using selective sensing.

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Theory, Methods and Applications of Active Learning
2:10	Learning to Discover
3:23	Laplace
4:06	What is Active Learning ?
6:40	Does active learning always help ?
8:06	Why Active Learning? Understanding the Mind (1)
8:42	Why Active Learning? Understanding the Mind (2)
10:09	Visual Perception (1)
11:39	Visual Perception (2)
12:08	Visual Perception (3)
13:32	Visual Perception (4)
14:06	Visual Perception (5)
16:11	Why Active Learning? Understanding Complex Systems
16:48	National Ecological Observation Network (NEON)
17:28	Learning by Queries
18:02	Why Active Learning? Automating Science
19:41	Machine Learning (Passive)
20:11	Active Learning
21:01	Robot Scientist
21:54	Hypothesis and Query/Feature Spaces
23:49	A Simple Algorithm for Noiseless Active Learning
26:30	Generalized Binary Search / Splitting Algorithm
27:57	Flavors of Active Learning and Analysis
29:46	What if there is noise or mismatch ?
33:45	Active Learning for Classification
34:00	What if there is noise or mismatch ?
36:48	Active Learning for Classification
37:59	Example
40:12	Active Learning for Regression
40:58	Active Learning for Image Processing
41:53	Active Learning for Fun !
42:41	Theoretical Foundations of Active Learning
44:04	Passive Learning
44:39	Semi-Supervised and Active Learning
47:01	Learning Rates and Sample Complexity
49:35	Research Questions
51:03	A few success stories
51:12	Learning a decision hyperplane in Rd (1)
53:27	Learning a decision hyperplane in Rd (2)
53:56	Now you see it, now you don't

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.