LVQ is a Gradient Based Maximum Margin Algorithm

Thread View: bne.seminar

1 messages

[T] Back to Threads [V] Tree View [G] Group Articles

1 total messages Started by Tony Rhodes Thu, 19 Aug 1999 00:00

#53

Author: Tony Rhodes
Date: Thu, 19 Aug 1999 00:00

27 lines
1568 bytes

Speaker: Mr. Lawrence Buckingham
	 PhD Student
	 School of Computing Science


Title: LVQ is a Gradient Based Maximum Margin Algorithm
	 
Date:    Tuesday, 24th August 1999 
Time:    12.00pm
Place:   Lecture Theatre S303, Faculty of Information Technology,
         Queensland University of Technology

Abstract:
---------
This seminar will describe an abstract decision surface estimation algorithm,
derived from Bayes' rule for classification.  The abstract algorithm
maximises a confidence measure (the margin) at each training point.  By
stipulating a specific functional form for the estimated probability
density of each class, a soft-competition supervised clustering
algorithm is produced.

The seminar will demonstrate that several concrete implementations of Learning 
Vector Quantization (LVQ) are approximate realisations of this algorithm, and 
thereby that LVQ is a gradient based algorithm which maximises the margin at 
each training point.
---

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads