Corpus ID: 13978804

Theory reconstruction: a representation learning view on predicate invention

@article{Dumancic2016TheoryRA,
  title={Theory reconstruction: a representation learning view on predicate invention},
  author={Sebastijan Dumancic and Wannes Meert and Hendrik Blockeel},
  journal={ArXiv},
  year={2016},
  volume={abs/1606.08660}
}
  • Sebastijan Dumancic, Wannes Meert, Hendrik Blockeel
  • Published in ArXiv 2016
  • Computer Science, Mathematics
  • 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. 
    7
    Twitter Mentions

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 20 REFERENCES

    Learning deep architectures for ai. Found

    • Y Bengio
    • Trends Mach. Learn
    • 2009
    VIEW 2 EXCERPTS

    Learning Deep Architectures for AI

    • Yoshua Bengio
    • Computer Science
    • Foundations and Trends in Machine Learning
    • 2007
    VIEW 1 EXCERPT

    Statistical predicate invention

    VIEW 2 EXCERPTS