Abstract
We investigate a variant of the on-line learning model for classes of \0,1\-valued functions (concepts) in which the labels of a certain amount of the input instances are corrupted by adversarial noise. We propose an extension of a general learning strategy, known as “Closure Algorithm”, to this noise model, and show a worst-case mistake bound of m + (d+1)K for learning an arbitrary intersection-closed concept class C, where K is the number of noisy labels, d is a combinatorial parameter measuring C's complexity, and m is the worst-case mistake bound of the Closure Algorithm for learning C in the noise-free model. For several concept classes our extended Closure Algorithm is efficient and can tolerate a noise rate up to the information-theoretic upper bound. Finally, we show how to efficiently turn any algorithm for the on-line noise model into a learning algorithm for the PAC model with malicious noise.
Similar content being viewed by others
References
D. Angluin, Queries and concept learning, Machine Learning 2(4) (1988) 319–342.
M. Anthony and J. Shawe-Taylor, A result of Vapnik with applications, Discrete Applied Mathematics 47 (1994) 207–217.
P. Auer, On-line learning of rectangles in noisy environments, in: Proceedings of the 6th Annual ACM Workshop on Computational Learning Theory (ACM Press, 1993) pp. 253–261.
P. Auer and P.M Long, Simulating access to hidden information while learning, in: Proceedings of the 26th ACM Symposium on the Theory of Computing (ACM Press, 1994) pp. 263–272.
S. Boucheron, Learnability from positive examples in the Valiant framework, Manuscript (1988).
N. Cesa-Bianchi, Models of learning with noise, Unpublished manuscript (1994).
N. Cesa-Bianchi, Y. Freund, D.P. Helmbold and M.K. Warmuth, On-line prediction and conversion strategies, in: Proceedings of the First Euro-COLT Workshop, The Institute of Mathematics and its Applications Conference Series – New Series Number 53 (Clarendon Press, Oxford, 1994) pp. 205–216.
Z. Chen and S. Homer, On learning counting functions with queries, in: Proceedings of the 7th Annual ACM Workshop on Computational Learning Theory (ACM Press, 1994) pp. 218–227.
P. Fischer and H.U. Simon, On learning ring-sum-expansions, SIAM Journal on Computing 21 (1992) 181–192.
D. Haussler, N. Littlestone and M.K. Warmuth, Predicting 0, 1-functions on randomly drawn points, Information and Computation 115(2) (1994) 248–292.
D.P. Helmbold and P.M Long, Tracking drifting concepts by minimizing disagreements, Machine Learning 14(1) (1994) 27–45.
D.P. Helmbold, R. Sloan and M.K. Warmuth, Learning nested differences of intersection-closed concept classes, Machine Learning 5(2) (1990) 165–196.
D.P. Helmbold, R. Sloan and M.K. Warmuth, Learning integer lattices, SIAM Journal on Computing 21(2) (1992) 240–266.
M.J. Kearns and M. Li, Learning in the presence of malicious errors, SIAM Journal on Computing 22(4) (1993) 807–837.
N. Littlestone, Learning quickly when irrelevant attributes abound: a new linear-threshold algorithm, Machine Learning 2(4) (1988) 285–318.
N. Littlestone, Mistake bounds and logarithmic linear-threshold learning algorithms, Ph.D. thesis, University of California at Santa Cruz (1989).
N. Littlestone and M.K. Warmuth, The weighted majority algorithm, Information and Computation 108 (1994) 212–261.
B.K. Natarajan, On learning boolean functions, in: Proceedings of the 19th ACM Symposium on the Theory of Computing (ACM Press, 1987) pp. 296–304.
B.K. Natarajan, Machine Learning: A Theoretical Approach (Morgan Kaufmann, San Mateo, CA, 1991).
A. Schrijver, Theory of Linear and Integer Programming (Wiley, New York, 1986).
L. Valiant, A theory of the learnable, Communications of the ACM 27(11) (1984) 1134–1142.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Auer, P., Cesa-Bianchi, N. On-line learning with malicious noise and the closure algorithm. Annals of Mathematics and Artificial Intelligence 23, 83–99 (1998). https://doi.org/10.1023/A:1018960107028
Issue Date:
DOI: https://doi.org/10.1023/A:1018960107028