See this document in CiteSeerX!

Noise Tolerant Learning Using Early Predictors (1999)  (Make Corrections)  (3 citations)
Shai Fine, Ran Gilad-Bachrach, Eli Shamir, Naftali Tishby



  Home/Search   Context   Related

 
View or download:
cs.huji.ac.il/labs/...nips99LT495.ps.gz
cs.huji.ac.il/~ranb...nips99LT495.ps.gz
Cached:  PS.gz  PS  PDF   Image  Update  Help

From:  cs.huji.ac.il/labs/learnin...shai (more)
Homepages:  S.Fine  

Rate this article: (best)
  Comment on this article  
(Enter summary)

Abstract: Generalization in most PAC learning analysis starts around O (d) examples, where d = V C dim of the class. Nevertheless, analysis of learning curves using statistical mechanics shows much earlier generalization [7]. Here we introduce a gadget called Early Predictor, which exists if somewhat better than random prediction of the label of an arbitrary instance can be obtained from labels of O (log d) random examples. We were able to show that by taking a majority vote over a committee of Early... (Update)

Context of citations to this paper:   More

...hence, this assumption may be regarded as questionable. However, in some cases, the MQ oracle can be e ciently simulated to meet our needs [8]. Throughout this paper we denote by the sample space and by H the hypotheses class. We endow H with a known probability measure p, and...

...hence, this assumption may be regarded as questionable. However, in some cases, the MQ oracle can be eciently simulated to meet our needs [52]. Throughout this chapter we denote by the sample space and by H the hypotheses class. We endow H with a known probability measure p, and...

Cited by:   More
A Universal Noise Cleaning Procedure via the Dual.. - Fine..   (Correct)
Knowledge Acquisition in Statistical Learning Theory - Fine (1999)   (Correct)
Noise Tolerant Learnability via the Dual Learning Problem - Fine, Gilad-Bachrach..   (Correct)

Active bibliography (related documents):   More   All
0.3:   On Using Extended Statistical Queries to Avoid Membership.. - Bshouty, Feldman (2002)   (Correct)
0.3:   Learning DNF Over The Uniform Distribution Using A Quantum.. - Bshouty, Jackson (1999)   (Correct)
0.0:   Geometric Concept Learning and Related Topics - Kwek (1997)   (Correct)

Similar documents based on text:   More   All
0.2:   Agnostic Classification of Markovian Sequences - El-Yaniv, Fine, Tishby (1998)   (Correct)
0.1:   Learning Using Query by Committee, Linear Separation.. - Fine, Gilad-Bachrach..   (Correct)
0.1:   The Strength of Weak Learnability - Schapire (1990)   (Correct)

Related documents from co-citation:   More   All
3:   Probabilistic methods in the geometry of Banach spaces (context) - Pisier - 1986
3:   Combinatorial Geometry (context) - Pach, Agarwal - 1995
3:   Learning from noisy examples (context) - Angluin, Laird - 1988

BibTeX entry:   (Update)

S. Fine, R. Gilad-Bachrach, E. Shamir, and N. Tishby. Noise tolerant learning using early predictors. subbmitted to NIPS 1999. http://citeseer.ist.psu.edu/fine99noise.html   More

@misc{ fine99noise,
  author = "S. Fine and R. Gilad-Bachrach and E. Shamir and N. Tishby",
  title = "Noise tolerant learning using early predictors",
  text = "S. Fine, R. Gilad-Bachrach, E. Shamir, and N. Tishby. Noise tolerant learning
    using early predictors. subbmitted to NIPS 1999.",
  year = "1999",
  url = "citeseer.ist.psu.edu/fine99noise.html" }
Citations (may not include all citations):
657   Bagging predictors - Breiman - 1996
509   A decision-theoretic generalization of on-line learning and .. - Freund, Schapire - 1997
243   Boosting in the margin: a new explanation for the effectiven.. - Schapire, Freund et al. - 1998
142   Learning from noisy examples (context) - Angluin, Laird - 1988
134   An experimental comparison of three methods for constructing.. - Dietterich - 1998
107   Efficient noise-tolerant learning from statistical queries - Kearns - 1993
81   Support vector method for function approximation - Vapnik, Golowich et al. - 1996
47   Rigorous learning curve bounds from statistical mechanics - Haussler, Kearns et al. - 1997
20   Hardness results for neural network approximation problems - Bartlett, Ben-David - 1999
15   Journal of Computer and System Sciences (context) - Helmbold, Warmuth et al. - 1995
3   Learning with queries corrupted by classification noise - Jackson, Shamir et al. - 1999
2   Unpublished manuscript (context) - Gofer, Shamir - 1998
1   Efficient Learning from Faulty Data (context) - Decator - 1995

Documents on the same site (http://www.cs.huji.ac.il/labs/learning/People/shai.html):   More
Noise Tolerant Learnability via the Dual Learning Problem - Fine, Gilad-Bachrach..   (Correct)
Agnostic Classification of Markovian Sequences - El-Yaniv, Fine, Tishby (1998)   (Correct)
Digit Recognition In Noisy Environments Via A Sequential.. - Fine, Saon, Gopinath (2002)   (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