OPEN HOUSE on Multi-Task and Complex Outputs Learning

Learning with structured inputs

author: Tong Zhang, Yahoo! Research, Yahoo! Research

Description

I will present a novel approach to semi-supervised learning that employs a method which we refer to as structural learning (aka multi-task learning). The idea is to learn predictive structures from many auxiliary problems that are created from the unlabeled data (and are related to the target problem), and then transfer the learned structure to the supervised target problem.

You might be experiencing some problems with Your Video player.

*Slides*
0:02	Learning with Structured Input
0:57	Outline
1:47	Supervised learning
1:54	Supervised learning
1:59	Semi-supervised learning problem
2:13	Chunking
3:45	Previous semi-supervised approaches to NLP
4:49	Our approach
5:50	Multi-task learning problem
6:09	Multi-task learning (structural learning)
6:24	Standard linear prediction model
7:40	Example input vector representation
8:10	New model for multi-task learning
10:10	Joint empirical risk minimization
11:16	Theoretical Justification
12:27	Alternating Structure Optimization Algorithm (ASO)
13:13	Semi-supervised learning
14:51	Alternating Structure Optimization Algorithm (ASO)
19:00	Semi-supervised learning
19:07	Create multiple tasks and their labeled data …
19:21	Semi-supervised learning method using multi-task structural learning

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.