Estimating Cartesian compression via deep learning

A. Lőrincz, András Sárkány, Zoltán Milacski, Zoltán Tősér

Research output: Conference contribution

2 Citations (Scopus)

Abstract

We introduce a learning architecture that can serve compression while it also satisfies the constraints of factored reinforcement learning. Our novel Cartesian factors enable one to decrease the number of variables being relevant for the ongoing task, an exponential gain in the size of the state space. We demonstrate the working, the limitations and the promises of the abstractions: we develop a representation of space in allothetic coordinates from egocentric observations and argue that the lower dimensional allothetic representation can be used for path planning. Our results on the learning of Cartesian factors indicate that (a) shallow autoencoders perform well in our numerical example and (b) if deeper networks are needed, e.g., for classification or regression, then sparsity should also be enforced at (some of) the intermediate layers.

Original languageEnglish
Title of host publicationArtificial General Intelligence - 9th International Conference, AGI 2016, Proceedings
PublisherSpringer Verlag
Pages294-304
Number of pages11
Volume9782
ISBN (Print)9783319416489
DOIs
Publication statusPublished - 2016
Event9th International Conference on Artificial General Intelligence, AGI 2016 - New York, United States
Duration: júl. 16 2016júl. 19 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9782
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other9th International Conference on Artificial General Intelligence, AGI 2016
CountryUnited States
CityNew York
Period7/16/167/19/16

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Fingerprint Dive into the research topics of 'Estimating Cartesian compression via deep learning'. Together they form a unique fingerprint.

  • Cite this

    Lőrincz, A., Sárkány, A., Milacski, Z., & Tősér, Z. (2016). Estimating Cartesian compression via deep learning. In Artificial General Intelligence - 9th International Conference, AGI 2016, Proceedings (Vol. 9782, pp. 294-304). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9782). Springer Verlag. https://doi.org/10.1007/978-3-319-41649-6_30