English

Theory reconstruction: a representation learning view on predicate invention

Machine Learning 2016-06-30 v2 Machine Learning Logic in Computer Science

Abstract

With this positional paper we present a representation learning view on predicate invention. The intention of this proposal is to bridge the relational and deep learning communities on the problem of predicate invention. We propose a theory reconstruction approach, a formalism that extends autoencoder approach to representation learning to the relational settings. Our intention is to start a discussion to define a unifying framework for predicate invention and theory revision.

Cite

@article{arxiv.1606.08660,
  title  = {Theory reconstruction: a representation learning view on predicate invention},
  author = {Sebastijan Dumancic and Wannes Meert and Hendrik Blockeel},
  journal= {arXiv preprint arXiv:1606.08660},
  year   = {2016}
}

Comments

3 pages, StaRAI 2016 submission

R2 v1 2026-06-22T14:36:36.620Z