Results 1  10
of
347
The strength of weak learnability
 Machine Learning
, 1990
"... Abstract. This paper addresses the problem of improving the accuracy of an hypothesis output by a learning algorithm in the distributionfree (PAC) learning model. A concept class is learnable (or strongly learnable) if, given access to a Source of examples of the unknown concept, the learner with h ..."
Abstract

Cited by 861 (24 self)
 Add to MetaCart
Abstract. This paper addresses the problem of improving the accuracy of an hypothesis output by a learning algorithm in the distributionfree (PAC) learning model. A concept class is learnable (or strongly learnable) if, given access to a Source of examples of the unknown concept, the learner with high probability is able to output an hypothesis that is correct on all but an arbitrarily small fraction of the instances. The concept class is weakly learnable if the learner can produce an hypothesis that performs only slightly better than random guessing. In this paper, it is shown that these two notions of learnability are equivalent. A method is described for converting a weak learning algorithm into one that achieves arbitrarily high accuracy. This construction may have practical applications as a tool for efficiently converting a mediocre learning algorithm into one that performs extremely well. In addition, the construction has some interesting theoretical consequences, including a set of general upper bounds on the complexity of any strong learning algorithm as a function of the allowed error e.
An Efficient Boosting Algorithm for Combining Preferences
, 1999
"... The problem of combining preferences arises in several applications, such as combining the results of different search engines. This work describes an efficient algorithm for combining multiple preferences. We first give a formal framework for the problem. We then describe and analyze a new boosting ..."
Abstract

Cited by 707 (18 self)
 Add to MetaCart
The problem of combining preferences arises in several applications, such as combining the results of different search engines. This work describes an efficient algorithm for combining multiple preferences. We first give a formal framework for the problem. We then describe and analyze a new boosting algorithm for combining preferences called RankBoost. We also describe an efficient implementation of the algorithm for certain natural cases. We discuss two experiments we carried out to assess the performance of RankBoost. In the first experiment, we used the algorithm to combine different WWW search strategies, each of which is a query expansion for a given domain. For this task, we compare the performance of RankBoost to the individual search strategies. The second experiment is a collaborativefiltering task for making movie recommendations. Here, we present results comparing RankBoost to nearestneighbor and regression algorithms.
An introduction to kernelbased learning algorithms
 IEEE TRANSACTIONS ON NEURAL NETWORKS
, 2001
"... This paper provides an introduction to support vector machines (SVMs), kernel Fisher discriminant analysis, and ..."
Abstract

Cited by 589 (54 self)
 Add to MetaCart
This paper provides an introduction to support vector machines (SVMs), kernel Fisher discriminant analysis, and
Boosting a Weak Learning Algorithm By Majority
, 1995
"... We present an algorithm for improving the accuracy of algorithms for learning binary concepts. The improvement is achieved by combining a large number of hypotheses, each of which is generated by training the given learning algorithm on a different set of examples. Our algorithm is based on ideas pr ..."
Abstract

Cited by 516 (15 self)
 Add to MetaCart
We present an algorithm for improving the accuracy of algorithms for learning binary concepts. The improvement is achieved by combining a large number of hypotheses, each of which is generated by training the given learning algorithm on a different set of examples. Our algorithm is based on ideas presented by Schapire in his paper "The strength of weak learnability", and represents an improvement over his results. The analysis of our algorithm provides general upper bounds on the resources required for learning in Valiant's polynomial PAC learning framework, which are the best general upper bounds known today. We show that the number of hypotheses that are combined by our algorithm is the smallest number possible. Other outcomes of our analysis are results regarding the representational power of threshold circuits, the relation between learnability and compression, and a method for parallelizing PAC learning algorithms. We provide extensions of our algorithms to cases in which the conc...
Principles and methods of Testing Finite State Machines  a survey
 PROCEEDINGS OF IEEE
, 1996
"... With advanced computer technology, systems are getting larger to fulfill more complicated tasks, however, they are also becoming less reliable. Consequently, testing is an indispensable part of system design and implementation; yet it has proved to be a formidable task for complex systems. This moti ..."
Abstract

Cited by 334 (14 self)
 Add to MetaCart
With advanced computer technology, systems are getting larger to fulfill more complicated tasks, however, they are also becoming less reliable. Consequently, testing is an indispensable part of system design and implementation; yet it has proved to be a formidable task for complex systems. This motivates the study of testing finite state machines to ensure the correct functioning of systems and to discover aspects of their behavior. A finite state machine contains a finite number of states and produces outputs on state transitions after receiving inputs. Finite state machines are widely used to model systems in diverse areas, including sequential circuits, certain types of programs, and, more recently, communication protocols. In a testing problem we have a machine about which we lack some information; we would like to deduce this information by providing a sequence of inputs to the machine and observing the outputs produced. Because of its practical importance and theoretical interest, the problem of testing finite state machines has been studied in different areas and at various times. The earliest published literature on this topic dates back to the 50’s. Activities in the 60’s and early 70’s were motivated mainly by automata theory and sequential circuit testing. The area seemed to have mostly died down until a few years ago when the testing problem was resurrected and is now being studied anew due to its applications to conformance testing of communication protocols. While some old problems which had been open for decades were resolved recently, new concepts and more intriguing problems from new applications emerge. We review the fundamental problems in testing finite state machines and techniques for solving these problems, tracing progress in the area from its inception to the present and the state of the art. In addition, we discuss extensions of finite state machines and some other topics related to testing.
Toward efficient agnostic learning
 In Proceedings of the Fifth Annual ACM Workshop on Computational Learning Theory
, 1992
"... Abstract. In this paper we initiate an investigation of generalizations of the Probably Approximately Correct (PAC) learning model that attempt to significantly weaken the target function assumptions. The ultimate goal in this direction is informally termed agnostic learning, in which we make virtua ..."
Abstract

Cited by 236 (8 self)
 Add to MetaCart
(Show Context)
Abstract. In this paper we initiate an investigation of generalizations of the Probably Approximately Correct (PAC) learning model that attempt to significantly weaken the target function assumptions. The ultimate goal in this direction is informally termed agnostic learning, in which we make virtually no assumptions on the target function. The name derives from the fact that as designers of learning algorithms, we give up the belief that Nature (as represented by the target function) has a simple or succinct explanation. We give a number of positive and negative results that provide an initial outline of the possibilities for agnostic learning. Our results include hardness results for the most obvious generalization of the PAC model to an agnostic setting, an efficient and general agnostic learning method based on dynamic programming, relationships between loss functions for agnostic learning, and an algorithm for a learning problem that involves hidden variables.
Training a 3Node Neural Network is NPComplete
, 1992
"... We consider a 2layer, 3node, ninput neural network whose nodes compute linear threshold functions of their inputs. We show that it is NPcomplete to decide whether there exist weights and thresholds for this network so that it produces output consistent with a given set of training examples. We ..."
Abstract

Cited by 227 (3 self)
 Add to MetaCart
(Show Context)
We consider a 2layer, 3node, ninput neural network whose nodes compute linear threshold functions of their inputs. We show that it is NPcomplete to decide whether there exist weights and thresholds for this network so that it produces output consistent with a given set of training examples. We extend the result to other simple networks. We also present a network for which training is hard but where switching to a more powerful representation makes training easier. These results suggest that those looking for perfect training algorithms cannot escape inherent computational difficulties just by considering only simple or very regular networks. They also suggest the importance, given a training problem, of finding an appropriate network and input encoding for that problem. It is left as an open problem to extend our result to nodes with nonlinear functions such as sigmoids.
Efficient Distributionfree Learning of Probabilistic Concepts
 Journal of Computer and System Sciences
, 1993
"... In this paper we investigate a new formal model of machine learning in which the concept (boolean function) to be learned may exhibit uncertain or probabilistic behaviorthus, the same input may sometimes be classified as a positive example and sometimes as a negative example. Such probabilistic c ..."
Abstract

Cited by 212 (8 self)
 Add to MetaCart
(Show Context)
In this paper we investigate a new formal model of machine learning in which the concept (boolean function) to be learned may exhibit uncertain or probabilistic behaviorthus, the same input may sometimes be classified as a positive example and sometimes as a negative example. Such probabilistic concepts (or pconcepts) may arise in situations such as weather prediction, where the measured variables and their accuracy are insufficient to determine the outcome with certainty. We adopt from the Valiant model of learning [27] the demands that learning algorithms be efficient and general in the sense that they perform well for a wide class of pconcepts and for any distribution over the domain. In addition to giving many efficient algorithms for learning natural classes of pconcepts, we study and develop in detail an underlying theory of learning pconcepts. 1 Introduction Consider the following scenarios: A meteorologist is attempting to predict tomorrow's weather as accurately as pos...
Inference of finite automata using homing sequences (Extended Abstract)
, 1989
"... We present new algorithms for inferring an unknown finitestate automaton from its input/output behavior in the absence of a means of re~rttinp the machine to a start date. A key technique used is inference of a homing sequence for the unknown automaton. Our infknrnce procedufes experiment with the ..."
Abstract

Cited by 193 (6 self)
 Add to MetaCart
We present new algorithms for inferring an unknown finitestate automaton from its input/output behavior in the absence of a means of re~rttinp the machine to a start date. A key technique used is inference of a homing sequence for the unknown automaton. Our infknrnce procedufes experiment with the unknown machine, and from time to time require a teacher to supply counterexamplea to incorrect conjectures about the structure of the unknown automaton. In this setting, we describe a learning algorithm which, with probability 1 6, outputs a correct deecription of the unknown machine in time polynomial in the automaton's size, the length of the longest counterexample, and log(ll6). We present an analogous algorithm which makes use of a diversitybased representation of the finitestate system. Our algorithms are the first which are provably eflecfive for these problems, in the absence oj Q &quot;reset.&quot; We also present probabilistic algorithms for permutation automata which do not require a teacher to supply counterexamples. For inferring a permutation automaton of diversity D, we improve the best previous time bound by roughly a factor of D³/ log D.