A note on batch and incremental learnability

DOI10.1006/JCSS.1998.1568MaRDI QIDQ1271611zbMATH OpenOpenAlexFDO

Authors A. K. Sharma

Publication date 10 November 1998

Published in Journal of Computer and System Sciences (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1006/jcss.1998.1568

zbMATH Keywords

recursively enumerable languages

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Review text

The author presents the result that the collections of recursively enumerable languages those are finite identified from both the positive and the negative data can also be identified in the limit only from the positive data by giving an algorithmic transformation on machines. This transformation also yields an algorithm to convert a batch learning system that learns from both positive and negative data into an incremental learning system that learns from the positive data only.

Recommendations

• Incremental learning from positive data
• On the power of incremental learning.
• Towards a Better Understanding of Incremental Learning
• Publication:4955305

Cites work

• Comparison of identification criteria for machine inductive inference
• scientific article; zbMATH DE number 3664335 (Why is no real title available?)
• scientific article; zbMATH DE number 408780 (Why is no real title available?)
• Language identification in the limit
• Toward a mathematical theory of inductive inference

Cited in

(5)

• Learning with refutation
• Some natural conditions on incremental learning
• Vacillatory and BC learning on noisy data
• On the learnability of vector spaces
• On ordinal VC-dimension and some notions of complexity

Describes a project that uses

Uses Software

• GOLEM

This page was built for publication: A note on batch and incremental learnability

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1271611)