Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
140 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Overdrawing Urns using Categories of Signed Probabilities (2312.12453v1)

Published 14 Dec 2023 in math.PR, cs.LG, and cs.LO

Abstract: A basic experiment in probability theory is drawing without replacement from an urn filled with multiple balls of different colours. Clearly, it is physically impossible to overdraw, that is, to draw more balls from the urn than it contains. This paper demonstrates that overdrawing does make sense mathematically, once we allow signed distributions with negative probabilities. A new (conservative) extension of the familiar hypergeometric ('draw-and-delete') distribution is introduced that allows draws of arbitrary sizes, including overdraws. The underlying theory makes use of the dual basis functions of the Bernstein polynomials, which play a prominent role in computer graphics. Negative probabilities are treated systematically in the framework of categorical probability and the central role of datastructures such as multisets and monads is emphasised.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (24)
  1. S. Abramsky & A. Brandenburger (2014): An Operational Interpretation of Negative Probabilities and No-Signalling Models. In F. van Breugel, E. Kashefi, C. Palamidessi & J. Rutten, editors: Horizons of the Mind. A Tribute to Prakash Panangaden, Lect. Notes Comp. Sci. 8464, Springer, Berlin, pp. 59–75, 10.1007/978-3-319-06880-0_3.
  2. A. Blass & Y. Gurevich (2021): Negative probabilities: what are they for? Journal of Physics A: Mathematical and Theoretical 54(31), p. 315303, 10.1088/1751-8121/abef4d.
  3. K. Cho & B. Jacobs (2019): Disintegration and Bayesian inversion via string diagrams. Math. Struct. in Comp. Sci. 29(7), pp. 938–971, 10.1017/s0960129518000488.
  4. In J. Esparza & A. Murawski, editors: Foundations of Software Science and Computation Structures, Lect. Notes Comp. Sci. 10203, Springer, Berlin, pp. 355–369, 10.1007/978-3-662-54458-7_21.
  5. P. Diaconis (1977): Finite forms of De Finetti’s theorem on exchangeability. Synthese 36(2), pp. 271–281, 10.1007/BF00486116.
  6. W. Dong-Bing (1993): Dual bases of a Bernstein polynomial basis on simplices. Computer aided geometric design 10(6), pp. 483–489, 10.1016/0167-8396(93)90025-X.
  7. R. Feynman (1987): Negative Probability. In B. Hiley & F. Peat, editors: Quantum Implications, Essays in Honor of David Bohm, Routledge and Kegan Paul, London, pp. 235–246, 10.1063/1.2811503.
  8. Advances in Math. 370, p. 107239, 10.1016/J.AIM.2020.107239.
  9. J. Hoschek & D. Lasser (1993): Fundamentals of computer aided geometric design. AK Peters, Ltd.
  10. B. Jacobs (2016): Affine Monads and Side-Effect-Freeness. In I. Hasuo, editor: Coalgebraic Methods in Computer Science (CMCS 2016), Lect. Notes Comp. Sci. 9608, Springer, Berlin, pp. 53–72, 10.1007/978-3-319-40370-0_5.
  11. B. Jacobs (2018): From Probability Monads to Commutative Effectuses. Journ. of Logical and Algebraic Methods in Programming 94, pp. 200–237, 10.1016/j.jlamp.2016.11.006.
  12. B. Jacobs (2020): A Channel-Based Perspective on Conjugate Priors. Math. Struct. in Comp. Sci. 30(1), pp. 44–61, 10.1017/S0960129519000082.
  13. B. Jacobs (2021): From Multisets over Distributions to Distributions over Multisets. In: Logic in Computer Science, IEEE, Computer Science Press, 10.1109/lics52264.2021.9470678.
  14. B. Jacobs (2022): Urns & Tubes. Compositionality 4(4), 10.32408/compositionality-4-4.
  15. B. Jacobs & S. Staton (2020): De Finetti’s construction as a categorical limit. In D. Petrişan & J. Rot, editors: Coalgebraic Methods in Computer Science (CMCS 2020), Lect. Notes Comp. Sci. 12094, Springer, Berlin, pp. 90–111, 10.1007/978-3-030-57201-3_6.
  16. E. Jaynes (1982): Some applications and extensions of the De Finetti representation theorem. Bayesian Inference and Decision Techniques with Applications: Essays in Honor of Bruno de Finetti. North-Holland Publishers.
  17. B. Jüttler (1998): The dual basis functions for the Bernstein polynomials. Adv. Computational Mathematics 8, pp. 345–352, 10.1023/A:1018912801267.
  18. G. Kerns & G. Székely (2006): De Finetti’s Theorem for abstract finite exchangeable sequences. Journal of Theoretical Probability 19(3), pp. 589–608, 10.1007/s10959-006-0028-z.
  19. G. Lorentz (2013): Bernstein polynomials. American Mathematical Soc.
  20. G. Meissner & M. Burgin (2011): Negative Probabilities in Financial Modeling. Wilmott Magazine, 10.2139/ssrn.1773077.
  21. P. Panangaden (2009): Labelled Markov Processes. Imperial College Press, London, 10.1142/p595.
  22. G. Székely (2005): Half of a Coin: Negative Probabilities. Wilmott Magazine, pp. 66–68.
  23. H. Tijms & K. Staats (2007): Negative probabilities at work in the M/D/1 queue. Probability in the Engineering and Informational Sciences 21(1), pp. 67–76, 10.1017/S0269964807070040.
  24. K. Zhao & J. Sun (1988): Dual bases of multivariate Bernstein-Bézier polynomials. Computer aided geometric design 5(2), pp. 119–125, 10.1016/0167-8396(88)90026-X.
Citations (1)

Summary

We haven't generated a summary for this paper yet.