Analogy-preserving Functions: A Way to Extend Boolean Samples

Abstract : Training set extension is an important issue in machine learning. Indeed when the examples at hand are in a limited quantity, the performances of standard classifiers may significantly decrease and it can be helpful to build additional examples. In this paper, we consider the use of analogical reasoning , and more particularly of analogical proportions for extending training sets. Here the ground truth labels are considered to be given by a (partially known) function. We examine the conditions that are required for such functions to ensure an error-free extension in a Boolean setting. To this end, we introduce the notion of Analogy Preserving (AP) functions, and we prove that their class is the class of affine Boolean functions. This noteworthy theoretical result is complemented with an empirical investigation of approximate AP functions, which suggests that they remain suitable for training set extension.
Document type :
Conference papers
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download

https://hal.inria.fr/hal-01668230
Contributor : Miguel Couceiro <>
Submitted on : Tuesday, December 19, 2017 - 9:58:13 PM
Last modification on : Friday, October 11, 2019 - 8:22:52 PM

File

0218.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01668230, version 1

Citation

Miguel Couceiro, Nicolas Hug, Henri Prade, Gilles Richard. Analogy-preserving Functions: A Way to Extend Boolean Samples. IJCAI 2017 - 26th International Joint Conference on Artificial Intelligence, Aug 2017, Melbourne, Australia. pp.1-7. ⟨hal-01668230⟩

Share

Metrics

Record views

257

Files downloads

159