Year |
Citation |
Score |
2019 |
De A, O’Donnell R, Servedio RA. Optimal mean-based algorithms for trace reconstruction Annals of Applied Probability. 29: 851-874. DOI: 10.1214/18-Aap1394 |
0.321 |
|
2019 |
Liu Z, Chen X, Servedio RA, Sheng Y, Xie J. Distribution-free Junta Testing Acm Transactions On Algorithms. 15: 1-23. DOI: 10.1145/3264434 |
0.385 |
|
2018 |
Chen X, Servedio RA, Tan L, Waingarten E, Xie J. Settling the Query Complexity of Non-adaptive Junta Testing Journal of the Acm. 65: 1-18. DOI: 10.1145/3213772 |
0.587 |
|
2018 |
De A, Servedio RA. A new central limit theorem and decomposition for Gaussian polynomials, with an application to deterministic approximate counting Probability Theory and Related Fields. 171: 981-1044. DOI: 10.1007/S00440-017-0804-Y |
0.48 |
|
2017 |
Chen X, Servedio RA, Tan L, Waingarten E, Xie J. Settling the Query Complexity of Non-Adaptive Junta Testing Electronic Colloquium On Computational Complexity. 24: 19. DOI: 10.4230/Lipics.Ccc.2017.26 |
0.586 |
|
2017 |
Håstad J, Rossman B, Servedio RA, Tan L. An Average-Case Depth Hierarchy Theorem for Boolean Circuits Journal of the Acm. 64: 35. DOI: 10.1145/3095799 |
0.597 |
|
2017 |
De A, Diakonikolas I, Servedio RA. The Inverse Shapley value problem Games and Economic Behavior. 105: 122-147. DOI: 10.1016/J.Geb.2017.06.004 |
0.366 |
|
2016 |
De A, Diakonikolas I, Servedio RA. A Robust Khintchine Inequality, and Algorithms for Computing Optimal Constants in Fourier Analysis and High-Dimensional Geometry Siam Journal On Discrete Mathematics. 30: 1058-1094. DOI: 10.1137/130919143 |
0.417 |
|
2015 |
Diakonikolas I, Jaiswal R, Servedio RA, Tan L, Wan A. Noise Stable Halfspaces are Close to Very Small Juntas Chicago Journal of Theoretical Computer Science. 2015: 1-13. DOI: 10.4086/Cjtcs.2015.004 |
0.566 |
|
2015 |
Rossman B, Servedio RA, Tan L. Complexity Theory Column 89: The Polynomial Hierarchy, Random Oracles, and Boolean Circuits Sigact News. 46: 50-68. DOI: 10.1145/2852040.2852052 |
0.573 |
|
2015 |
Canonne CL, Ron D, Servedio RA. Testing probability distributions using conditional samples Siam Journal On Computing. 44: 540-616. DOI: 10.1137/130945508 |
0.347 |
|
2015 |
Daskalakis C, Diakonikolas I, Servedio RA. Learning Poisson Binomial Distributions Algorithmica. 72: 316-357. DOI: 10.1007/S00453-015-9971-3 |
0.487 |
|
2015 |
Ron D, Servedio RA. Exponentially Improved Algorithms and Lower Bounds for Testing Signed Majorities Algorithmica. 72: 400-429. DOI: 10.1007/S00453-013-9858-0 |
0.508 |
|
2014 |
Daskalakis C, Diakonikolas I, Servedio RA. Learning k-Modal Distributions via Testing Theory of Computing. 10: 535-570. DOI: 10.4086/Toc.2014.V010A020 |
0.374 |
|
2014 |
Diakonikolas I, Servedio RA, Tan L, Wan A. A Regularity Lemma and Low-Weight Approximators for Low-Degree Polynomial Threshold Functions Theory of Computing. 10: 27-53. DOI: 10.4086/Toc.2014.V010A002 |
0.569 |
|
2014 |
De A, Diakonikolas I, Feldman V, Servedio RA. Nearly optimal solutions for the chow parameters problem and low-weight approximation of halfspaces Journal of the Acm. 61. DOI: 10.1145/2590772 |
0.73 |
|
2014 |
Long PM, Servedio RA. On the weight of halfspaces over hamming balls Siam Journal On Discrete Mathematics. 28: 1035-1061. DOI: 10.1137/120868402 |
0.383 |
|
2014 |
Diakonikolas I, Raghavendra P, Servedio RA, Tan LY. Average sensitivity and noise sensitivity of polynomial threshold functions Siam Journal On Computing. 43: 231-253. DOI: 10.1137/110855223 |
0.637 |
|
2013 |
De A, Diakonikolas I, Servedio RA. Deterministic Approximate Counting for Degree-2 Polynomial Threshold Functions. Electronic Colloquium On Computational Complexity. 20: 172. DOI: 10.14288/1.0043373 |
0.462 |
|
2013 |
Diakonikolas I, Servedio RA. Improved Approximation of Linear Threshold Functions Computational Complexity. 22: 623-677. DOI: 10.1007/S00037-012-0045-5 |
0.449 |
|
2011 |
Gopalan P, O'Donnell R, Servedio RA, Shpilka A, Wimmer K. Testing fourier dimensionality and sparsity Siam Journal On Computing. 40: 1075-1100. DOI: 10.1137/100785429 |
0.472 |
|
2011 |
Jackson JC, Lee HK, Servedio RA, Wan A. Learning random monotone DNF Discrete Applied Mathematics. 159: 259-271. DOI: 10.1016/J.Dam.2010.08.022 |
0.664 |
|
2011 |
Diakonikolas I, Lee HK, Matulef K, Servedio RA, Wan A. Efficiently testing sparse GF(2) polynomials Algorithmica (New York). 61: 580-605. DOI: 10.1007/S00453-010-9426-9 |
0.659 |
|
2010 |
Diakonikolas I, Gopalan P, Jaiswal R, Servedio RA, Viola E. Bounded independence fools halfspaces Siam Journal On Computing. 39: 3441-3462. DOI: 10.1137/100783030 |
0.406 |
|
2010 |
Matulef K, O'Donnell R, Rubinfeld R, Servedio RA. Testing Halfspaces Siam Journal On Computing. 39: 2004-2047. DOI: 10.1137/070707890 |
0.463 |
|
2010 |
Long PM, Servedio RA. Random classification noise defeats all convex potential boosters Machine Learning. 78: 287-304. DOI: 10.1007/S10994-009-5165-Z |
0.393 |
|
2010 |
O’Donnell R, Servedio RA. New degree bounds for polynomial threshold functions Combinatorica. 30: 327-358. DOI: 10.1007/S00493-010-2173-3 |
0.46 |
|
2009 |
Glasner D, Servedio RA. Distribution-Free Testing Lower Bound for Basic Boolean Functions Theory of Computing. 5: 191-216. DOI: 10.4086/Toc.2009.V005A010 |
0.329 |
|
2009 |
Rubinfeld R, Servedio RA. Testing monotone high-dimensional distributions Random Structures and Algorithms. 34: 24-44. DOI: 10.1002/Rsa.V34:1 |
0.358 |
|
2008 |
Dachman-Soled D, Lee HK, Malkin T, Servedio RA, Wan A, Wee H. Optimal cryptographic hardness of learning monotone functions Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 5125: 36-47. DOI: 10.4086/Toc.2009.V005A013 |
0.654 |
|
2008 |
O'Donnell R, Servedio RA. The Chow parameters problem Proceedings of the Annual Acm Symposium On Theory of Computing. 517-526. DOI: 10.1137/090756466 |
0.538 |
|
2008 |
Feldman J, O'Donnell R, Servedio RA. Learning Mixtures of Product Distributions over Discrete Domains Siam Journal On Computing. 37: 1536-1564. DOI: 10.1137/060670705 |
0.501 |
|
2008 |
Kalai AT, Klivans AR, Mansour Y, Servedio RA. Agnostically Learning Halfspaces Siam Journal On Computing. 37: 1777-1805. DOI: 10.1137/060649057 |
0.498 |
|
2008 |
Atici A, Servedio RA. Learning unions of ω (1)-dimensional rectangles Theoretical Computer Science. 405: 209-222. DOI: 10.1016/J.Tcs.2008.06.036 |
0.682 |
|
2007 |
O'Donnell R, Servedio RA. Learning Monotone Decision Trees in Polynomial Time Siam Journal On Computing. 37: 827-844. DOI: 10.1137/060669309 |
0.458 |
|
2007 |
Long PM, Servedio RA, Simon HU. Discriminative learning can succeed where generative learning fails Information Processing Letters. 103: 131-135. DOI: 10.1016/J.Ipl.2007.03.004 |
0.437 |
|
2007 |
Atici A, Servedio RA. Quantum algorithms for learning and testing juntas Quantum Information Processing. 6: 323-348. DOI: 10.1007/S11128-007-0061-6 |
0.67 |
|
2007 |
Lee HK, Servedio RA, Wan A. DNF are teachable in the average case Machine Learning. 69: 79-96. DOI: 10.1007/S10994-007-5007-9 |
0.594 |
|
2007 |
Servedio RA. Every Linear Threshold Function has a Low-Weight Approximator Computational Complexity. 16: 180-209. DOI: 10.1007/S00037-007-0228-7 |
0.496 |
|
2006 |
Jackson JC, Servedio RA. On Learning Random DNF Formulas Under the Uniform Distribution Theory of Computing. 2: 147-172. DOI: 10.4086/Toc.2006.V002A008 |
0.387 |
|
2006 |
Arias M, Feigelson A, Khardon R, Servedio RA. Polynomial certificates for propositional classes Information and Computation. 204: 816-834. DOI: 10.1016/J.Ic.2006.03.001 |
0.427 |
|
2005 |
Khardon R, Roth D, Servedio RA. Efficiency versus convergence of Boolean kernels for on-line learning algorithms Journal of Artificial Intelligence Research. 24: 341-356. DOI: 10.1613/Jair.1655 |
0.705 |
|
2005 |
Jackson JC, Servedio RA. Learning Random Log-Depth Decision Trees under Uniform Distribution Siam Journal On Computing. 34: 1107-1128. DOI: 10.1137/S0097539704444555 |
0.43 |
|
2005 |
Servedio RA, Wan A. Computing sparse permanents faster Information Processing Letters. 96: 89-92. DOI: 10.1016/J.Ipl.2005.06.007 |
0.394 |
|
2005 |
Atici A, Servedio RA. Improved bounds on quantum learning algorithms Quantum Information Processing. 4: 355-386. DOI: 10.1007/S11128-005-0001-2 |
0.655 |
|
2005 |
Khardon R, Servedio RA. Maximum Margin Algorithms with Boolean Kernels Journal of Machine Learning Research. 6: 1405-1429. DOI: 10.1007/978-3-540-45167-9_8 |
0.512 |
|
2004 |
Servedio RA, Gortler SJ. Equivalences and Separations Between Quantum and Classical Learnability Siam Journal On Computing. 33: 1067-1092. DOI: 10.1137/S0097539704412910 |
0.425 |
|
2004 |
Klivans AR, Servedio RA. Learning intersections of halfspaces with a margin Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science). 3120: 348-362. DOI: 10.1016/J.Jcss.2007.04.012 |
0.492 |
|
2004 |
Mossel E, O'Donnell R, Servedio RA. Learning functions of k relevant variables Journal of Computer and System Sciences. 69: 421-434. DOI: 10.1016/J.Jcss.2004.04.002 |
0.516 |
|
2004 |
Klivans AR, Servedio RA. Learning DNF in time 2Õ(n(1/3)) Journal of Computer and System Sciences. 68: 303-318. DOI: 10.1016/J.Jcss.2003.07.007 |
0.505 |
|
2004 |
Servedio RA. On learning monotone DNF under product distributions Information & Computation. 193: 57-74. DOI: 10.1016/J.Ic.2004.04.003 |
0.481 |
|
2004 |
Servedio RA. Monotone Boolean formulas can approximate monotone linear threshold functions Discrete Applied Mathematics. 142: 181-187. DOI: 10.1016/J.Dam.2004.02.003 |
0.393 |
|
2003 |
Servedio RA. Smooth boosting and learning with malicious noise Journal of Machine Learning Research. 4: 633-648. DOI: 10.1162/153244304773936072 |
0.448 |
|
2003 |
Klivans AR, Servedio RA. Boosting and hard-core set construction Machine Learning. 51: 217-238. DOI: 10.1023/A:1022949332276 |
0.443 |
|
2003 |
Kalai A, Servedio RA. Boosting in the presence of noise Conference Proceedings of the Annual Acm Symposium On Theory of Computing. 196-205. DOI: 10.1016/J.Jcss.2004.10.015 |
0.441 |
|
2002 |
Servedio RA. Perceptron, Winnow, and PAC Learning Siam Journal On Computing. 31: 1358-1369. DOI: 10.1137/S0097539798340928 |
0.51 |
|
2002 |
Servedio R. PAC Analogues of Perceptron and Winnow Via Boosting the Margin Machine Learning. 47: 133-151. DOI: 10.1023/A:1013633619373 |
0.465 |
|
2000 |
Servedio RA. Computational Sample Complexity and Attribute-Efficient Learning Journal of Computer and System Sciences. 60: 161-178. DOI: 10.1006/Jcss.1999.1666 |
0.452 |
|
Show low-probability matches. |