AA1*: A Dynamic Incremental Network that Learns by Discrimination

Giraud‐Carrier, Christophe; Martinez, Tony

doi:10.1007/978-3-7091-7535-4_14

Cited by 2 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AA1 learns by discrimination and constructs a network in which knowledge is distributed across all of the nodes. In a previous paper, we presented AA1 , an extension of AA1 that focuses on minimizing the network's growth to improve predictive accuracy and making lower-order features available for discrimination during Node Selection [14]. Here, we formalize and fill in many of the details missing from AA1 's earlier account.…”

Section: Introductionmentioning

confidence: 99%

Learning by Discrimination: A Constructive Incremental Approach

Giraud‐Carrier¹,

Martinez²

2007

JCP

Self Cite

View full text Add to dashboard Cite

This paper presents i-AA1*, a constructive, incremental learning algorithm for a special class of weightless, self-organizing networks. In i-AA1*, learning consists of adapting the nodes’ functions and the network’s overall topology as each new training pattern is presented. Provided the training data is consistent, computational complexity is low and prior factual knowledge may be used to “prime” the network and improve its predictive accuracy and/or efficiency. Empirical generalization results on both toy problems and more realistic tasks demonstrate promise.

show abstract

Section: Introductionmentioning

confidence: 99%

Learning by Discrimination: A Constructive Incremental Approach

Giraud‐Carrier¹,

Martinez²

2007

JCP

Self Cite

View full text Add to dashboard Cite

show abstract

“…AA1 learns by discrimination and constructs a network in which knowledge is distributed across all of the nodes. In a previous paper, we presented AA1 , an extension of AA1 that focuses on minimizing the network's growth to improve predictive accuracy and making lower-order features available for discrimination during Node Selection [19]. Here, we make an explicit assumption of consistency, thus enabling true incremental learning [20], and demonstrate the algorithm's ability to exploit prior factual knowledge in learning.…”

Section: Introductionmentioning

confidence: 99%

A Constructive Incremental Learning Algorithm for Binary Classification Tasks

Giraud‐Carrier

Martinez

2006

2006 IEEE Mountain Workshop on Adaptive and Learning Systems

Self Cite

View full text Add to dashboard Cite

This paper presents i-AA1 , a constructive, incremental learning algorithm for a special class of weightless, selforganizing networks. In i-AA1 , learning consists of adapting the nodes' functions and the network's overall topology as each new training pattern is presented. Provided the training data is consistent, computational complexity is low and prior factual knowledge may be used to "prime" the network and improve its predictive accuracy. Empirical generalization results on both toy problems and more realistic tasks demonstrate promise.

show abstract

AA1*: A Dynamic Incremental Network that Learns by Discrimination

Cited by 2 publications

References 3 publications

Learning by Discrimination: A Constructive Incremental Approach

Learning by Discrimination: A Constructive Incremental Approach

A Constructive Incremental Learning Algorithm for Binary Classification Tasks

Contact Info

Product

Resources

About