Prahladh Harsha
Adam Klivans
Skip Abstract Section
Abstract
Let X be randomly chosen from {-1,1}n, and let Y be randomly chosen from the standard spherical Gaussian on ℝn. For any (possibly unbounded) polytope P formed by the intersection of k halfspaces, we prove that |Pr[X ∈ P] - Pr[Y ∈ P]| ≤ log8/5k ⋅ Δ, where Δ is a parameter that is small for polytopes formed by the intersection of “regular” halfspaces (i.e., halfspaces with low influence). The novelty of our invariance principle is the polylogarithmic dependence on k. Previously, only bounds that were at least linear in k were known. The proof of the invariance principle is based on a generalization of the Lindeberg method for proving central limit theorems and could be of use elsewhere.
We give two important applications of our invariance principle, one from learning theory and the other from pseudorandomness.
(1) A bound of logO(1)k ⋅ ϵ1/6 on the Boolean noise sensitivity of intersections of k “regular” halfspaces (previous work gave bounds linear in k). This gives a corresponding agnostic learning algorithm for intersections of regular halfspaces.
(2) A pseudorandom generator (PRG) for estimating the Gaussian volume of polytopes with k faces within error δ and seed-length O(log n poly(log k,1/δ)).
We also obtain PRGs with similar parameters that fool polytopes formed by intersection of regular halfspaces over the hypercube. Using our PRG constructions, we obtain the first deterministic quasi-polynomial time algorithms for approximately counting the number of solutions to a broad class of integer programs, including dense covering problems and contingency tables.
- Ailon, N. and Chazelle, B. 2006. Approximate nearest neighbors and the fast Johnson-Lindenstrauss transform. In Proceedings of the 38th ACM Symposium on Theory of Computing (STOC). 557--563. Google Scholar
Digital Library
- Austrin, P. 2007. Balanced max 2-sat might not be the hardest. In Proceedings of the 39th ACM Symposium on Theory of Computing (STOC), D. S. Johnson and U. Feige, Eds. ACM, 189--197. http://doi.acm.org.10.1145 1250790.1250818. Google ScholarDigital Library
- Austrin, P. 2010. Towards sharp inapproximability for any 2-CSP. SIAM J. Comput 39, 6, 2430--2463. http://dx.doi.org/10.1137/07011670. Google ScholarDigital Library
- Bansal, N. and Khot, S. 2010. Inapproximability of hypergraph vertex cover and applications to scheduling problems. In Proceedings of the International Colloquim on Automata, Languages, and Programming (ICALP), S. Abramsky, C. Gavoille, C. Kirchner, F. M. auf der Heide, and P. G. Spirakis, Eds. Lecture Notes in Computer Science Series, vol. 6198. Springer, 250--261. http://dx.doi.org/10.1007/978-3-642-14165-2. Google ScholarDigital Library
- Barvinok, A. and Veomett, E. 2008. The computational complexity of convex bodies. In Surveys on Discrete and Computational Geometry: Twenty Years Later, J. E. Goodman, J. Pach, and R. Pollack, Eds., Contemporary Mathematics Series, vol. 453. AMS, 117--137. http://www.ams.org/bookstore-getitem/item=COMM-453, arXiv:math/0610325.Google Scholar
- Beck, M. and Robins, S. 2007. Computing the Continuous Discretely: Integer-point Enumeration in Polyhedra 1st Ed. Undergraduate Texts in Mathematics, Springer. http://math.sfsu.edu/beck/ccd.html.Google Scholar
- Benjamini, I., Kalai, G., and Schramm, O. 1999. Noise sensitivity of Boolean functions and applications to percolation. Inst. Hautes Études Sci. Publ. Math. 90, 1, 5--43.Google ScholarCross Ref
- Bentkus, V. K. 1990. Smooth approximations of the norm and differentiable functions with bounded support in Banach space lk∞. Lithuan. Math. J. 30, 3, 223--230.Google ScholarCross Ref
- Bentkus, V. K. 2003. On the dependence of the Berry-Esseen bound on dimension. J. Stat. Plan. Inf. 113, 2, 385--402.Google ScholarCross Ref
- Blais, E. and O'Donnell, R. 2010. Lower bounds for testing function isomorphism. In Proceedings of the IEEE Conference on Computational Complexity. IEEE Computer Society, 235--246. http://doi. ieeecomputersociety.org/10.1109/CCC.2010.30. Google ScholarDigital Library
- Chatterjee, S. 2005. A simple invariance theorem. arXiv:math/0508213.Google Scholar
- Cryan, M. and Dyer, M. E. 2003. A polynomial-time algorithm to approximately count contingency tables when the number of rows is constant. J. Comput. Syst. Sci. 67, 2, 291--310. Google ScholarDigital Library
- de Wolf, R. 2008. A brief introduction to Fourier analysis on the Boolean cube. Theory of Comput. Graduate Sur. 1, 1--20.Google ScholarCross Ref
- Diakonikolas, I., Harsha, P., Klivans, A., Meka, R., Raghavendra, P., Servedio, R. A., and Tan, L.-Y. 2010a. Bounding the average sensitivity and noise sensitivity of polynomial threshold functions. In Proceedings of the 42nd ACM Symposium on Theory of Computing (STOC), ACM, 533--542. http://doi.acm.org/10.1145/1806689.1806763. Google ScholarDigital Library
- Diakonikolas, I., Kane, D. M., and Nelson, J. 2010b. Bounded independence fools degree-2 threshold functions. In Proceedings of the 51st IEEE Symposium on Foundations of Computer Science (FOCS). IEEE, 11--20. Google ScholarDigital Library
- Dinur, I., Friedgut, E., and Regev, O. 2008. Independent sets in graph powers are almost contained in juntas. Geomet. Funct. Anal. 18, 1, 77--97.Google ScholarCross Ref
- Dinur, I., Mossel, E., and Regev, O. 2009. Conditional hardness for approximate coloring. SIAM J. Comput 39, 3, 843--873. http://dx.doi.org/10.1137/07068062X. Google ScholarDigital Library
- Dyer, M. E. 2003. Approximate counting by dynamic programming. In Proceedings of the 35th ACM Symposium on Theory of Computing (STOC). 693--699. Google ScholarDigital Library
- Feldman, V., Guruswami, V., Raghavendra, P., and Wu, Y. 2009. Agnostic learning of monomials by halfspaces is hard. In Proceedings of the 50th IEEE Symposium on Foundation of Computer Science (FOCS). IEEE Computer Society, 385--394. http://doi.ieeecomputersociety.org/10.1109/FOCS.2009.26. Google ScholarDigital Library
- Feller, W. 1968. An Introduction to Probability Theory and Its Applications, Volume 1 3rd Ed. Wiley.Google Scholar
- Feller, W. 1971. An Introduction to Probability Theory and Its Applications, Volume 2 2nd Ed. Wiley.Google Scholar
- Goemans, M. X. and Williamson, D. P. 1995. Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming. J. ACM 42, 6, 1115--1145. Google ScholarDigital Library
- Gopalan, P., Klivans, A., and Meka, R. 2010a. Polynomial-time approximation schemes for knapsack and related counting problems using branching programs. Electron. Colloq. Comput. Complex. (ECCC) 17, 133. http://eccc.hpi-web.de/report/2010/133.Google Scholar
- Gopalan, P., O'Donnell, R., Wu, Y., and Zuckerman, D. 2010b. Fooling functions of halfspaces under product distributions. In Proceedings of 25th IEEE Conference on Computational Complexity 223--234. Google ScholarDigital Library
- Harsha, P., Klivans, A., and Meka, R. 2010. An invariance principle for polytopes. In Proceedings of the 42nd ACM Symposium on Theory of Computing (STOC). ACM, 543--552. Google ScholarDigital Library
- Håstad, J. 2001. Some optimal inapproximability results. J. ACM 48, 4, 798--859. Google ScholarDigital Library
- Haussler, D. 1992. Decision theoretic generalizations of the PAC model for neural net and other learning applications. Inf. Comput. 100, 1, 78--150. Google ScholarDigital Library
- Impagliazzo, R., Nisan, N., and Wigderson, A. 1994. Pseudorandomness for network algorithms. In Proceedings of the 26th ACM Symposium on Theory of Computing (STOC). ACM, 356--364. Google ScholarDigital Library
- Jerrum, M. and Sinclair, A. 1997. The Markov chain Monte Carlo method: An approach to approximate counting and integration. In Approximation Algorithms for NP-hard Problems, D. S. Hochbaum, Ed., PWS Publishing Company. http://www.ieor.berkeley.edu/-hochbaum/html/book-aanp.html. Google ScholarDigital Library
- Kahn, J., Kalai, G., and Linial, N. 1988. The influence of variables on Boolean functions (extended abstract). In Proceedings of the 29th IEEE Symposium on Foundations of Computer Science (FOCS). 68--80. Google ScholarDigital Library
- Kalai, A. T., Klivans, A. R., Mansour, Y., and Servedio, R. A. 2008. Agnostically learning halfspaces. SIAM J. Comput. 37, 6, 1777--1805. Google ScholarDigital Library
- Kalai, G. 2005. Noise sensitivity and chaos in social choice theory. Tech. rep. 399, Center for Rationality and Interactive Decision Theory, Hebrew University of Jerusalem. http://www.ratio.huji.ac.il/dp_files/dp-399.pdf.Google Scholar
- Kearns, M. J., Schapire, R. E., and Sellie, L. 1994. Toward efficient agnostic learning. Mach. Learn. 17, 2--3, 115--141. Google ScholarDigital Library
- Khot, S., Kindler, G., Mossel, E., and O'Donnell, R. 2007. Optimal inapproximability results for MAX-CUT and other 2-variable CSPs? SIAM J. Comput. 37, 1, 319--357. Google ScholarDigital Library
- Klivans, A. R., O'Donnell, R., and Servedio, R. A. 2004. Learning intersections and thresholds of halfspaces. J. Comput. Syst. Sci. 68, 4, 808--840. Google ScholarDigital Library
- Klivans, A. R., O'Donnell, R., and Servedio, R. A. 2008. Learning geometric concepts via Gaussian surface area. In Proceedings of the Symposium on Foundations of Computer Science (FOCS) 541--550. Google ScholarDigital Library
- Linial, N., Mansour, Y., and Nisan, N. 1993. Constant depth circuits, Fourier transform, and learnability. J. ACM 40, 3, 607--620. Google ScholarDigital Library
- Mahajan, S. and Hariharan, R. 1999. Derandomizing approximation algorithms based on semidefinite programming. SIAM J. Comput. 28, 5, 1641--1663. Google ScholarDigital Library
- Mansour, Y. 1994. Learning Boolean functions via the Fourier transform. In Theoretical Advances in Neural Computation and Learning, V. P. Roychowdhury, K.-Y. Siu, and A. Orlitsky, Eds., Kluwer Academic Publishers, 391--424. http://www.springer.com/physics/complexity/book/978-0-7923-9478-5.Google Scholar
- Meka, R. and Zuckerman, D. 2010. Pseudorandom generators for polynomial threshold functions. In Proceedings of the ACM Symposium on Theory of Computing (STOC), 427--436. Google ScholarDigital Library
- Mossel, E. 2008. Gaussian bounds for noise correlation of functions and tight analysis of long codes. In Proceedings of the Symposium on Foundations of Computer Science (FOCS), 156--165. Google ScholarDigital Library
- Mossel, E. 2012. A quantitative arrow theorem. Probab. Theory Relat. Fields 154, 49--88. DOI: 10.1007/s00440-011-0362-7.Google ScholarCross Ref
- Mossel, E., O'Donnell, R., and Oleszkiewicz, K. 2005. Noise stability of functions with low influences invariance and optimality. In Proceedings of the 46th IEEE Symposium on Foundations of Computer Science (FOCS). 21--30. Google ScholarDigital Library
- Naor, J. and Naor, M. 1993. Small-bias probability spaces: Efficient constructions and applications. SIAM J. Comput. 22, 4, 838--856. Google ScholarDigital Library
- Nazarov, F. 2003. On the maximal perimeter of a convex set in ℝn with respect to a Gaussian measure. In Geometric Aspects of Functional Analysis (Israel Seminar 2001--2002). Lecture Notes in Mathematics Series, vol. 1807/2003, Springer, 169--187.Google Scholar
- O'Donnell, R. 2004. Hardness amplification within NP. J. Comput. Syst. Sci. 69, 1, 68--94. Google ScholarDigital Library
- O'Donnell, R. 2008. Some topics in analysis of Boolean functions. In Proceedings of the 40th ACM Symposium on Theory of Computing (STOC). 569--578. Google ScholarDigital Library
- O'Donnell, R. and Wu, Y. 2009. Conditional hardness for satisfiable 3-CSPs. In Proceedings of the 41st ACM Symposium on Theory of Computing (STOC). 493--502. http://doi.acm.org/10.1145/1536414.1536482. Google ScholarDigital Library
- Paulauskas, V. and Račkauskas, A. 1989. Approximation Theory in the Central Limit Theorem: Exact Results in Banach Spaces. Kluwer Academic Publishers. (Translated from Russian).Google Scholar
- Peres, Y. 2004. Noise stability of weighted majority. (arXiv:math/0412377.)Google Scholar
- Pinelis, I. 1994. Extremal probabilistic problems and hotelling T2 test under a symmetry condition. Ann. Statist. 22, 1, 357--368.Google ScholarCross Ref
- Raghavendra, P. 2008. Optimal algorithms and inapproximability results for every CSP? In STOC, ACM, 245--254. http://doi.acm.org/10.1145/1374376. Google ScholarDigital Library
- Rotar', V. I. 1979. Limit theorems for polylinear forms. J. Multivar. Anal. 9, 4, 511--530.Google ScholarCross Ref
- Shi, Y. 2000. Lower bounds of quantum black-box complexity and degree of approximating polynomials by influence of Boolean variables. Inf. Process. Lett. 75, 1-2, 79--83. Google ScholarDigital Library
- Tao, T. 2009. Talagrand's concentration inequality. (Post in Blog “What's new”). htt://terrytao.wordpress. com/2009/06/09/talagrands-concentration-inequality/.Google Scholar
- Wolff, P. 2007. Hypercontractivity of simple random variables. Studia Math 180, 3, 219--236.Google ScholarCross Ref
- Ziegler, G. M. 1995. Lectures on polytopes. Graduate texts in Mathematics Series, vol. 152. Springer. http://www.springer.com/math/geometry/book/978-0-387-94365-7.Google Scholar
Index Terms
- An invariance principle for polytopes
Recommendations
An invariance principle for polytopes
STOC '10: Proceedings of the forty-second ACM symposium on Theory of computingLet X be randomly chosen from {-1,1}n, and let Y be randomly chosen from the standard spherical Gaussian on Rn. For any (possibly unbounded) polytope P formed by the intersection of k halfspaces, we prove that |Pr[X ∈ P] - Pr[Y ∈ P]| ≤ log8/5k • Δ, ...
The complexity of constructing pseudorandom generators from hard functions
We study the complexity of constructing pseudorandom generators (PRGs) from hard functions, focussing on constant-depth circuits. We show that, starting from a function f : {0,1}l --> {0,1} computable in alternating time O(l) with O(1) alternations that ...
Average-case hardness of NP from exponential worst-case hardness assumptions
STOC 2021: Proceedings of the 53rd Annual ACM SIGACT Symposium on Theory of ComputingA long-standing and central open question in the theory of average-case complexity is to base average-case hardness of NP on worst-case hardness of NP. A frontier question along this line is to prove that PH is hard on average if UP requires (sub-)...
Comments