Pareto-optimal data compression for binary classification tasks

@article{Tegmark2019ParetooptimalDC,
  title={Pareto-optimal data compression for binary classification tasks},
  author={Max Tegmark and Tailin Wu},
  journal={ArXiv},
  year={2019},
  volume={abs/1908.08961}
}
The goal of lossy data compression is to reduce the storage cost of a data set $X$ while retaining as much information as possible about something ($Y$) that you care about. For example, what aspects of an image $X$ contain the most information about whether it depicts a cat? Mathematically, this corresponds to finding a mapping $X\to Z\equiv f(X)$ that maximizes the mutual information $I(Z,Y)$ while the entropy $H(Z)$ is kept below some fixed threshold. We present a method for mapping out the… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-3 OF 3 CITATIONS

References

Publications referenced by this paper.
SHOWING 1-10 OF 26 REFERENCES

Tech

  • A. Krizhevsky, G. Hinton
  • Rep., Citeseer
  • 2009
VIEW 9 EXCERPTS
HIGHLY INFLUENTIAL

Neural computation 29

  • D. Strouse, D. J. Schwab
  • 1611
  • 2017
VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Entropy 21

  • T. Wu, I. Fischer, I. Chuang, M. Tegmark
  • 924
  • 2019
VIEW 3 EXCERPTS

Neural computation 31

  • D. Strouse, D. J. Schwab
  • 596
  • 2019
VIEW 2 EXCERPTS

and D

  • A. Tan, L. Meshulam, W. Bialek
  • Schwab, Bulletin of the American Physical Society
  • 2019
VIEW 1 EXCERPT

and K

  • D. G. Clark, J. A. Livezey
  • E. Bouchard, arXiv preprint arXiv:1905.09944
  • 2019
VIEW 1 EXCERPT

and M

  • T. Wu, I. Fischer, I. Chuang
  • Tegmark, arXiv preprint arXiv:1907.07331
  • 2019
VIEW 3 EXCERPTS

Physical Review X 8

  • J. Thompson, A. J. Garner, +3 authors M. Gu
  • 031013
  • 2018
VIEW 1 EXCERPT

The Conditional Entropy Bottleneck