Shai Ben-David Nadav Eiron Department of Computer Science Technion Haifa...  Home/Search
Context Related
| cs.technion.ac.il/~sha...philhard.ps.gz Cached: PS.gz PS PDF This document uses CoBlitz to cache paper downloads. If your firewall is blocking outgoing connections to port 3125, you can use these links to download local copies.
Image Update HelpPS.gz PS PDF From: cs.technion.ac.il/~shai/pub (more) (Enter author homepages) |
Rate this article:      (best)Comment on this article |
Abstract: We address the computational complexity of learning in the agnostic framework. For a variety of common concept classes we prove that, unless P=NP, there is no polynomial time approximation scheme for finding a member in the class that approximately maximizes the agreement with a given training sample. In particular our results apply to the classes of monomials, axis-aligned hyper-rectangles, closed balls and monotone monomials. For each of these classes we prove the NP-hardness of approximating ... (Update)
Context of citations to this paper: More
...P6=NP, for any ffi 1=22, there is no polynomial time ffi approximation algorithm for MAX E2 SAT. Theorem 2. 3 [Ben David, Eiron and Long, [1]] Assuming P6=NP, for any ffi 3=418, there is no polynomial time ffi approximation algorithm for BSH. Claim 2.4 BSH cp pol DOH....
.... many interesting classes F , for small enough fi (in the order of 0:005) unless P=NP there exist no fi weak agnostic learner ( 2] [3]) However, no currently known result rules out the existence of agnostic fi weak learners for these classes once fi is sufficiently large...
Cited by: More
The Computational Complexity of Densest Region Detection - Ben-David, Eiron, Simon (2000) (Correct)
Efficient Learning of Linear Perceptrons - Ben-David, Simon (2000) (Correct)
Hardness Results for Neural Network Approximation Problems - Bartlett, Ben-David (1999) (Correct)
Active bibliography (related documents): More All
0.9: On the Difficulty of Approximately Maximizing Agreements - Ben-David, Eiron, Long (2000) (Correct)
0.1: Theory and Applications of Agnostic PAC-Learning with Small .. - Auer, Holte, Maass (1995) (Correct)
0.1: Theory and Applications of Agnostic PAC-learning with Small.. - Auer (1995) (Correct)
Similar documents based on text: More All
0.1: Limitations of Learning Via Embeddings in Euclidean Half.. - Ben-David, Eiron, Simon (2002) (Correct)
0.1: Agnostic Boosting - Ben-David, Long, Mansour (Correct)
0.1: Limitations of Learning Via Embeddings in Euclidean Half-Spaces - Ben-David, Eiron (Correct)
Related documents from co-citation: More All
3: the complexity of polyhedral separability - Megiddo - 1988
3: Efficient learning of linear perceptrons - Ben-David, Simon - 2000
2: Some optimal inapproximability results - Hastad - 1997
BibTeX entry: (Update)
Shai Ben-David, Nadav Eiron, and Philip Long. On the difficulty of approximately maximizing agreements. Submitted to this conference. http://citeseer.ist.psu.edu/455606.html More
@inproceedings{ bendavid00difficulty,
author = "Shai Ben-David and Nadav Eiron and Philip M. Long",
title = "On the Difficulty of Approximately Maximizing Agreements",
booktitle = "Proc. 13th Annu. Conference on Comput. Learning Theory",
publisher = "Morgan Kaufmann, San Francisco",
pages = "266--274",
year = "2000",
url = "citeseer.ist.psu.edu/455606.html" }
Citations (may not include all citations):509 A decision-theoretic generalization of on-line learning and .. - Freund, Schapire - 1995
500 Experiments with a new boosting algorithm - Freund, Schapire - 1996
273 The strength of weak learnability - Schapire - 1990
268 Decision theoretic generalizations of the PAC model for neur.. (context) - Haussler - 1992
226 for approximating set cover (context) - Feige, of - 1996
184 Cryptographic limitations on learning boolean formulae and f.. - Kearns, Valiant - 1994
144 Computational limitations on learning from examples (context) - Pitt, Valiant - 1988
142 Learning from noisy examples (context) - Angluin, Laird - 1988
120 Some optimal inapproximability results (context) - Hastad - 1997
120 Hardness of approximate optima in lattices (context) - Arora, Babai et al. - 1997
102 Training a 3-node neural network is NP-complete - Blum, Rivest - 1992
102 Toward efficient agnostic learning - Kearns, Schapire et al. - 1994
94 Learning in the presence of malicious errors - Kearns, Li - 1993
90 Neural Network Design and the Complexity of Learning (context) - Judd - 1990
48 The complexity and approximability of finding maximum feasib.. - Amaldi, Kann - 1995
37 Tracking drifting concepts by minimizing disagreements - Helmbold, Long - 1994
31 Cryptographic hardness of distribution specific learning (context) - Kharitonov - 1993
20 Hardness results for neural network approximation problems - Bartlett, Ben-David - 1999
20 the complexity of training neural networks with continuous a.. - DasGupta, Siegelmann et al. - 1995
10 Robust trainability of single neurons (context) - Hoffgen, Simon et al. - 1995
7 with applications in computer graphics and machine learning (context) - Dobkin, Gunopulos et al. - 1996
2 Preliminary version in STOC (context) - Blumer, Ehrenfeucht et al. - 1989
1 the hardness of unsupervised learning (context) - Ben-David, Eiron et al. - 2000
Documents on the same site (http://www.cs.technion.ac.il/~shai/pub.html): More
Combinatorial Variability of Vapnik-Chervonenkis Classes.. - Ben-David, Litman (Correct)
Scale-sensitive Dimensions, Uniform Convergence, and.. - Alon, Ben-David (1997) (Correct)
Agnostic Boosting - Ben-David, Long, Mansour (Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC