Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 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

On Confidence Sequences for Bounded Random Processes via Universal Gambling Strategies (2207.12382v3)

Published 25 Jul 2022 in math.PR, cs.IT, math.IT, and stat.ME

Abstract: This paper considers the problem of constructing a confidence sequence, which is a sequence of confidence intervals that hold uniformly over time, for estimating the mean of bounded real-valued random processes. This paper revisits the gambling-based approach established in the recent literature from a natural \emph{two-horse race} perspective, and demonstrates new properties of the resulting algorithm induced by Cover (1991)'s universal portfolio. The main result of this paper is a new algorithm based on a mixture of lower bounds, which closely approximates the performance of Cover's universal portfolio with constant per-round time complexity. A higher-order generalization of a lower bound on a logarithmic function in (Fan et al., 2015), which is developed as a key technique for the proposed algorithm, may be of independent interest.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (33)
  1. Thomas M Cover. Behavior of sequential predictors of binary sequences. In Trans. Fourth Prague Conf. Inf. Theory, September 1966.
  2. Thomas M Cover. Universal portfolios. Math. Financ., 1(1):1–29, 1991.
  3. Universal portfolios with side information. IEEE Trans. Inf. Theory, 42(2):348–363, 1996.
  4. Elements of information theory. John Wiley & Sons, 2006.
  5. Confidence sequences for mean, variance, and median. Proc. Natl. Acad. Sci. U. S. A., 58(1):66, 1967.
  6. Exponential inequalities for martingales with applications. Electron. J. Probab., 20:1–22, 2015.
  7. Minimum description length revisited. Int. J. Ind. Math., 11(01):1930001, 2019.
  8. Safe testing. In Proc. UCSD Inf. Theory Appl. Workshop, February 2020.
  9. Peter D Grünwald. The minimum description length principle. MIT press, 2007.
  10. Harrie Hendriks. Test martingales for bounded random variables. arXiv preprint arXiv:1801.09418, arXiv:2109.08923, 2021.
  11. Wassily Hoeffding. Probability inequalities for sums of bounded random variables. J. Am. Stat. Assoc., 58(301):13–30, 1963.
  12. Time-uniform, nonparametric, nonasymptotic confidence sequences. Ann. Statist., 49(2):1055–1080, 2021.
  13. Parameter-free online convex optimization with sub-exponential noise. In Conf. Learn. Theory, pages 1802–1823. PMLR, 2019.
  14. John L Kelly. A new interpretation of information rate. Bell Syst. Tech. J., 35(4):917–926, 1956.
  15. The performance of universal encoding. IEEE Trans. Inf. Theory, 27(2):199–207, 1981.
  16. Tze Leung Lai. On confidence sequences. Ann. Statist., 4(2):265–280, 1976.
  17. Efficient online portfolio with logarithmic regret. In Adv. Neural Inf. Proc. Syst., volume 31, 2018.
  18. Damped online Newton step for portfolio selection. arXiv preprint arXiv:2202.07574, 2022.
  19. Tight concentrations and confidence sequences from the regret of universal portfolio. IEEE Trans. Inf. Theory, 70(1):436–455, 2024. doi: 10.1109/TIT.2023.3330187. arXiv:2110.14099.
  20. Coin betting and parameter-free online learning. In Adv. Neural Inf. Proc. Syst., 2016.
  21. On equivalence of martingale tail bounds and deterministic regret inequalities. In Satyen Kale and Ohad Shamir, editors, Conf. Learn. Theory, volume 65 of Proceedings of Machine Learning Research, pages 1704–1722. PMLR, 2017.
  22. Admissible anytime-valid sequential inference must rely on nonnegative martingales. arXiv preprint arXiv:2009.03167, September 2020.
  23. Herbert Robbins. Statistical methods related to the law of the iterated logarithm. Ann. Math. Stat., 41(5):1397–1409, 1970.
  24. Glenn Shafer. Testing by betting: A strategy for statistical and scientific communication. J. R. Stat. Soc. Ser. A Stat. Soc., 184(2):407–431, April 2021.
  25. Game-theoretic foundations for probability and finance. Wiley Series in Probability and Statistics. John Wiley & Sons, Nashville, TN, March 2019.
  26. Test martingales, Bayes factors and p-values. Stat. Sci., 26(1):84–101, February 2011.
  27. Justin Solomon. Numerical algorithms: methods for computer vision, machine learning, and graphics. CRC press, 2015.
  28. Open problem: Fast and optimal online portfolio selection. In Conf. Learn. Theory, pages 3864–3869. PMLR, 2020.
  29. Jean Ville. Etude critique de la notion de collectif. Bull. Amer. Math. Soc, 45(11):824, 1939.
  30. SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python. Nature Methods, 17:261–272, 2020. doi: 10.1038/s41592-019-0686-2.
  31. Confidence sequences for sampling without replacement. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin, editors, Adv. Neural Inf. Proc. Syst., volume 33, pages 20204–20214. Curran Associates, Inc., 2020a. URL https://proceedings.neurips.cc/paper/2020/file/e96c7de8f6390b1e6c71556e4e0a4959-Paper.pdf.
  32. Estimating means of bounded random variables by betting. arXiv preprint arXiv:2010.09686, 2020b.
  33. Pushing the efficiency-regret Pareto frontier for online learning of portfolios and quantum states. arXiv preprint arXiv:2202.02765, 2022.
Citations (4)
View on Semantic Scholar

Summary

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

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Github Logo Streamline Icon: https://streamlinehq.com

GitHub

  1. GitHub - jongharyu/confidence-sequence-via-gambling (1 star)