Dale Schuurmans

Dynamic Weighted Majority: An Ensemble Method for Drifting Concepts ∗ (2009)

While methods for comparing two learning algorithms on a single data set have been scrutinized for quite some time already, the issue of statistical tests for comparisons of more algorithms on...

J. Zico Kolter, Marcus A. Maloof, Dale Schuurmans

Dual Representations for Dynamic Programming (2008)

Tao Wang, Daniel Lizotte, Michael Bowling, Dale Schuurmans

We propose a dual approach to dynamic programming and reinforcement learning, based on maintaining an explicit representation of visit distributions as opposed to value functions. An advantage of...

Stochastic Analysis of Lexical and Semantic Enhanced Structural Language Model (2008)

Shaojun Wang, Shaomin Wang, Russell Greiner, Dale Schuurmans

Abstract. In this paper, we present a directed Markov random field model that integrates trigram models, structural language models (SLM) and probabilistic latent semantic analysis (PLSA) for the...

Shape Time Discriminative Classification of Video Objects (2008)

Li Cheng, Baochun Bai, Cheng Lei, Dale Schuurmans, Shaojun Wang

We propose a discriminative approach to non-rigid video objects classification. Our goal is to recognize actions of the objects that appear in a video sequence, based on its shape time dynamics. This...

Stable Dual Dynamic Programming (2008)

Tao Wang, Daniel Lizotte, Michael Bowling, Dale Schuurmans

Recently, we have introduced a novel approach to dynamic programming and reinforcement learning that is based on maintaining explicit representations of stationary distributions instead of value...

c ○ 2002 Kluwer Academic Publishers. Manufactured in The Netherlands. Metric-Based Methods for Adaptive Model Selection (2008)

An Online Discriminative Approach to Background Subtraction (2008)

Abstract. We present a general approach to model selection and regularization that exploits unlabeled data to adaptively control hypothesis complexity in supervised learning tasks. The idea is to...

Li Cheng, Shaojun Wang, Dale Schuurmans, Terry Caelli

We present a simple, principled approach to detecting foreground objects in video sequences in real-time. Our method is based on an on-line discriminative learning technique that is able to cope with...

Minimax Regret Classifier for Imprecise Class Distributions (2008)

May 22, 2006 22:28 WSPC/Trim Size: 11in x 8.5in for Proceedings csb2006 Protein Fold Recognition Using the Gradient Boost Algorithm (2008)

The design of a minimum risk classifier based on data usually stems from the stationarity assumption that the conditions during training and test are the same: the misclassification costs assumed...

Feng Jiao, Jinbo Xu, Libo Yu, Dale Schuurmans

Protein structure prediction is one of the most important and difficult problems in computational molecular biology. Protein threading represents one of the most promising techniques for this...

What’s Strange About Recent Events (WSARE): An Algorithm for the Early Detection of Disease Outbreaks (2008)

Weng-keen Wong, Andrew Moore, Gregory Cooper, Michael Wagner, Dale Schuurmans

Traditional biosurveillance algorithms detect disease outbreaks by looking for peaks in a univariate time series of health-care data. Current health-care surveillance data, however, are no longer...

c ○ 2002 Kluwer Academic Publishers. Manufactured in The Netherlands. Metric-Based Methods for Adaptive Model Selection (2008)

Abstract. We present a general approach to model selection and regularization that exploits unlabeled data to adaptively control hypothesis complexity in supervised learning tasks. The idea is to...

Abstract (2008)

Discriminative Batch Mode Active Learning (2008)

We present a generalized view of support vector machines that does not rely on a Euclidean geometric interpretation nor even positive semidefinite kernels. We base our development instead on the...

Dale Schuurmans, Dana Wilkinson, Yuhong Guo

Active learning sequentially selects unlabeled instances to label with the goal of reducing the effort needed to learn a good classifier. Most previous studies in active learning have focused on...

Finnegan Southey (2008)

Discriminative Batch Mode Active Learning (2008)

Discriminative Batch Mode Active Learning (2008)

Active learning sequentially selects unlabeled instances to label with the goal of reducing the effort needed to learn a good classifier. Most previous studies in active learning have focused on...

Abstract (2008)

Li Cheng, Shaojun Wang, Dale Schuurmans, Terry Caelli

We present two new algorithms for online learning in reproducing kernel Hilbert spaces. Our first algorithm, ILK (implicit online learning with kernels), employs a new, implicit update technique that...

Constraint-based Optimization and Utility Elicitation using the Minimax Decision Criterion (2008)

Active learning sequentially selects unlabeled instances to label with the goal of reducing the effort needed to learn a good classifier. Most previous studies in active learning have focused on...

Craig Boutilier A, Relu Patrascu A, Pascal Poupart B, Dale Schuurmans

In many situations, a set of hard constraints encodes the feasible configurations of some system or product over which multiple users have distinct preferences. However, making suitable decisions...

Stable Dual Dynamic Programming (2008)

Tao Wang, Daniel Lizotte, Michael Bowling, Dale Schuurmans

Recently, we have introduced a novel approach to dynamic programming and reinforcement learning that is based on maintaining explicit representations of stationary distributions instead of value...

Stable Dual Dynamic Programming (2008)

Tao Wang, Daniel Lizotte, Michael Bowling, Dale Schuurmans

Recently, we have introduced a novel approach to dynamic programming and reinforcement learning that is based on maintaining explicit representations of stationary distributions instead of value...

Compact, Convex Upper Bound Iteration for Approximate POMDP Planning (2008)

Tao Wang, Pascal Poupart, Michael Bowling, Dale Schuurmans

Partially observable Markov decision processes (POMDPs) are an intuitive and general way to model sequential decision making problems under uncertainty.

Session Boundary Detection for Association (2008)

Rule Learning Using, Xiangji Huang, Fuchun Peng, Aijun An, Dale Schuurmans, Nick Cercone

We present a statistical method using n-gram language models to identify session boundaries in a large collection of Livelink log data.

Automatic Basis Selection Techniques for RBF Networks (2008)

Ali Ghodsi, Dale Schuurmans

This paper proposes a generic criterion that defines the optimum number of basis functions for radial basis function neural networks. The generalization performance of an RBF network relates to its...

Semi-supervised convex training for dependency parsing (2008)

Qin Iris Wang, Dale Schuurmans, Dekang Lin

We present a novel semi-supervised training algorithm for learning dependency parsers. By combining a supervised large margin loss with an unsupervised least squares loss, a discriminative, convex,...

Efficient global optimization for exponential family PCA and low-rank matrix factorization (2008)

Efficient exploration for optimizing immediate reward (2007)

Abstract—We present an efficient global optimization algorithm for exponential family principal component analysis (PCA) and associated low-rank matrix factorization problems. Exponential family...

Characterizing the Benefits of Model-Based Vs. Direct-Control Learning in Exploration (2007)

We consider the problem of learning an effective behavior strategy from reward. Although much studied, the issue of how to use prior knowledge to scale optimal behavior learning up to real-world...

Characterizing the Benefits of Model-Based Vs. Direct-Control Learning in Exploration (2007)

: We consider the problem of learning a good control function from reinforcement. Although much studied, the issue of how (or even whether) to use prior knowledge to scale reinforcement learning up...

Greedy importance sampling: A new Monte Carlo inference method (2007)

: We consider the problem of learning a good control function from reinforcement. Although much studied, the issue of how (or even whether) to use prior knowledge to scale reinforcement learning up...

Abstract Waterloo at NTCIR-3: Using Self-supervised Word Segmentation (2007)

This paper presents a new Monte Carlo inference method that is a simple variation of importance sampling. It is well known that importance sampling fails when the proposal distribution concentrates...

1 (2007)

Fuchun Peng, Xiangji Huang, Dale Schuurmans, Nick Cercone, Stephen Robertson

We propose a self-supervised word-segmentation technique for Chinese information retrieval. This method combines the advantages of traditional dictionary based approaches with character based...

z (2007)

Fuchun Peng, Dale Schuurmans, Vlado Keselj, Shaojun Wang

We present a method for computerassisted authorship attribution based on character-level n-gram language models. Our approach is based on simple information theoretic principles, and achieves...

Xiangji Huang, Fuchun Peng, Dale Schuurmans, Nick Cercone

In this paper, we describe the system we use

2 (2007)

Xiangji Huang, Fuchun Peng, Aijun An, Dale Schuurmans, Nick Cercone

Abstract. We present a statistical method using n-gram language models to identify session boundaries in a large collection of Livelink log data. The identied sessions are then used for association...

Waterloo at NTCIR-3: Using Self-supervised Word Segmentation (2007)

Xiangji Huang, Fuchun Peng, Dale Schuurmans, Nick Cercone

In this paper, we describe the system we use in the NTCIR-3 CLIR (cross language IR) task. We participate the SLIR (single language IR) track. In our system, we use a self-supervised...

y (2007)

Fuchun Peng, Xiangji Huang, Dale Schuurmans, Shaojun Wang

We present a simple approach for Asian language text classification without word segmentation, based on statistical n-gram language modeling. In particular, we examine Chinese and Japanese text...

Boltzmann Machine Learning with the Latent Maximum Entropy Principle (2007)

Shaojun Wang, Dale Schuurmans, Fuchun Peng, Yunxin Zhao

We present a new statistical learning paradigm for Boltzmann machines based on a new inference principle we have proposed: the latent maximum entropy principle (LME). LME is dierent both from...

y (2007)

Shaojun Wang, Ronald Rosenfeld, Yunxin Zhao, Dale Schuurmans

z

A Simple Closed-Class/Open-Class Factorization for Improved Language Modeling (2007)

The Sparse Data Problem in Statistical Language Modeling and Unsupervised Word Segmentation (2007)

We describe a simple improvement to n-gram language models where we estimate the distribution over closed-class (function) words separately from the conditional distribution of open-class words given...

Fuchun Peng, Supervisor Prof, Dale Schuurmans, Prof Frank Tompa

The sparse data problem is one of the most important problems in natural language processing. In this thesis, we are focusing on the sparse data problem in statistical language modeling and...

Efficient exploration for optimizing immediate reward (2007)

Text Classification in Asian Languages without Word Segmentation (2007)

We consider the problem of learning an effective behavior strategy from reward. Although much studied, the issue of how to use prior knowledge to scale optimal behavior learning up to real-world...

Fuchun Peng, Fuchun Schuurmans, Dale Schuurmans, Xiangji Huang, Shaojun Wang

We present a simple approach for Asian language text classification without word segmentation, based on statistical language modeling. In particular, we examine Chinese and Japanese text...

Automatic Complexity Control for System Identification (2007)

Ali Ghodsi, Dale Schuurmans

As a prerequisite for system identi cation based on c-mean clustering (FCM), it is necessary to assign the number of underlying partitions to be used for a given data set. However, for the FCM...

Automatic basis selection for RBF networks using Stein's unbiased risk estimator (2007)

Ali Ghodsi, Dale Schuurmans

The problem of selecting the appropriate number of basis functions is a critical issue for radial basis function neural networks. An RBF network with an overly restricted basis gives poor predictions...

Boltzmann Machine Learning with the Latent Maximum Entropy Principle (2007)

Shaojun Wang, Dale Schuurmans, Fuchun Peng, Yunxin Zhao

We present a new statistical learning paradigm for Boltzmann machines based on a new inference principle we have proposed: the latent maximum entropy principle (LME). LME is dierent both from...

Automatic gait optimization with gaussian process regression (2007)

Daniel Lizotte, Tao Wang, Michael Bowling, Dale Schuurmans

Gait optimization is a basic yet challenging problem for both quadrupedal and bipedal robots. Although techniques for automating the process exist, most involve local function optimization procedures...

Automatic gait optimization with gaussian process regression (2007)

Daniel Lizotte, Tao Wang, Michael Bowling, Dale Schuurmans

Gait optimization is a basic yet challenging problem for both quadrupedal and bipedal robots. Although techniques for automating the process exist, most involve local function optimization procedures...

Automatic gait optimization with gaussian process regression (2007)

Daniel Lizotte, Tao Wang, Michael Bowling, Dale Schuurmans

Gait optimization is a basic yet challenging problem for both quadrupedal and bipedal robots. Although techniques for automating the process exist, most involve local function optimization procedures...

Learning gene regulatory networks via globally regularized risk minimization (2007)

Learning gene regulatory networks via globally regularized risk minimization (2007)

Abstract. Learning the structure of a gene regulatory network from time-series gene expression data is a significant challenge. Most approaches proposed in the literature to date attempt to predict...

Convex relaxations of latent variable training (2007)

Abstract. Learning the structure of a gene regulatory network from time-series gene expression data is a significant challenge. Most approaches proposed in the literature to date attempt to predict...

Convex relaxations of latent variable training (2007)

We investigate a new, convex relaxation of an expectation-maximization (EM) variant that approximates a standard objective while eliminating local minima. First, a cautionary result is presented,...

Automatic gait optimization with gaussian process regression (2007)

We investigate a new, convex relaxation of an expectation-maximization (EM) variant that approximates a standard objective while eliminating local minima. First, a cautionary result is presented,...

Daniel Lizotte, Tao Wang, Michael Bowling, Dale Schuurmans

Gait optimization is a basic yet challenging problem for both quadrupedal and bipedal robots. Although techniques for automating the process exist, most involve local function optimization procedures...

Convex relaxations of latent variable training (2007)

Graphical models and point pattern matching (2006)

We investigate a new, convex relaxation of an expectation-maximization (EM) variant that approximates a standard objective while eliminating local minima. First, a cautionary result is presented,...

Tibério S. Caetano, Terry Caelli, Dale Schuurmans

Abstract—This paper describes a novel solution to the rigid point pattern matching problem in Euclidean spaces of any dimension. Although we assume rigid motion, jitter is allowed. We present a...

Web communities identification from random walks (2006)

Jiayuan Huang, Tingshao Zhu, Dale Schuurmans

Abstract. We propose a technique for identifying latent Web communities based solely on the hyperlink structure of the WWW, via random walks. Although the topology of the Directed Web Graph encodes...

Discriminative unsupervised learning of structured predictors (2006)

Linli Xu, Dana Wilkinson, Dale Schuurmans

We present a new unsupervised algorithm for training structured predictors that is discriminative, convex, and avoids the use of EM. The idea is to formulate an unsupervised version of structured...

Discriminative unsupervised learning of structured predictors (2006)

Linli Xu, Dana Wilkinson, Dale Schuurmans

We present a new unsupervised algorithm for training structured predictors that is discriminative, convex, and avoids the use of EM. The idea is to formulate an unsupervised version of structured...

Learning to model spatial dependency: Semi-supervised discriminative random fields (2006)

Chi-hoon Lee, Feng Jiao, Shaojun Wang, Dale Schuurmans, Russell Greiner

We present a novel, semi-supervised approach to training discriminative random fields (DRFs) that efficiently exploits labeled and unlabeled training data to achieve improved accuracy in a variety of...

Discriminative unsupervised learning of structured predictors (2006)

Linli Xu, Dana Wilkinson, Dale Schuurmans

We present a new unsupervised algorithm for training structured predictors that is discriminative, convex, and avoids the use of EM. The idea is to formulate an unsupervised version of structured...

Information marginalization on subgraphs (2006)

Jiayuan Huang, Russell Greiner, Dengyong Zhou, Dale Schuurmans

Abstract. Real-world data often involves objects that exhibit multiple relationships; for example, ‘papers ’ and ‘authors ’ exhibit both paperauthor interactions and paper-paper citation...

Learning to model spatial dependency: Semi-supervised discriminative random fields (2006)

Chi-hoon Lee, Feng Jiao, Shaojun Wang, Dale Schuurmans, Russell Greiner

We present a new, semi-supervised extension of discriminative random fields (DRFs) that efficiently exploits labeled and unlabeled training data to achieve improved accuracy in a variety of image...

Information marginalization on subgraphs (2006)

Jiayuan Huang, Tingshao Zhu, Russell Greiner, Dale Schuurmans, Dengyong Zhou

Real-world data often involve objects that exhibit multiple relations. A typical learning problem requires one to make inferences about a subclass of objects, while using the remaining objects and...

Information marginalization on subgraphs (2006)

Jiayuan Huang, Tingshao Zhu, Russell Greiner, Dengyong Zhou, Dale Schuurmans

Abstract. Real-world data often involves objects that exhibit multiple relationships; for example, ‘papers ’ and ‘authors ’ exhibit both paperauthor interactions and paper-paper citation...

Learning coordination classifiers (2005)

Yuhong Guo, Russell Greiner, Dale Schuurmans

We present a new approach to ensemble classification that requires learning only a single base classifier. The idea is to learn a classifier that simultaneously predicts pairs of test labels—as...

DOI: 10.1007/s10994-005-1505-9 Combined SVM-Based Feature Selection and Classification (2005)

Julia Neumann, Christoph Schnörr, Dale Schuurmans

Abstract. Feature selection is an important combinatorial optimisation problem in the context of supervised pattern classification. This paper presents four novel continuous feature selection...

Exploiting Syntactic, Semantic and Lexical Regularities in Language Modeling via Directed Markov Random Fields (2005)

Shaojun Wang, Shaomin Wang, Russell Greiner, Dale Schuurmans, Li Cheng

We present a directed Markov random field (MRF) model that combines n-gram models, probahilistic context free grammars (l'C FGs) and probabilistic latent semantic analysis (PLSA) for the purpose...

Maximum margin clustering (2005)

Linli Xu, James Neufeld, Bryce Larson, Dale Schuurmans

We propose a new method for clustering based on finding maximum margin hyperplanes through data. By reformulating the problem in terms of the implied equivalence relation matrix, we can pose the...

Strictly lexical dependency parsing (2005)

Qin Iris Wang, Dale Schuurmans, Dekang Lin

We present a strictly lexical parsing model where all the parameters are based on the words. This model does not rely on part-of-speech tags or grammatical categories. It maximizes the conditional...

Combining Statistical Language Models via the Latent Maximum Entropy Principle (2005)

Shaojun Wang Swang, Shaojun Wang, Dale Schuurmans, Fuchun Peng, Yunxin Zhao

In this paper, we present a unified probabilistic framework for statistical language modeling which can simultaneously incorporate various aspects of natural language, such as local word interaction,...

Exploiting Syntactic, Semantic and Lexical Regularities in Language Modeling via Directed Markov Random Fields (2005)

Shaojun Wang, Shaomin Wang, Russell Greiner, Dale Schuurmans, Li Cheng

We present a directed Markov random field (MRF) model that combines ¢-gram models, probabilistic context free grammars (PCFGs) and probabilistic latent semantic analysis (PLSA) for the purpose of...

Learning coordination classifiers (2005)

Yuhong Guo, Russell Greiner, Dale Schuurmans

We present a new approach to ensemble classification that requires learning only a single base classifier. The idea is to learn a classifier that simultaneously predicts pairs of test labels—as...

Variational bayesian image modelling (2005)

Li Cheng, Feng Jiao, Dale Schuurmans, Shaojun Wang

We present a variational Bayesian framework for performing inference, density estimation and model selection in a special class of graphical models—Hidden Markov Random Fields (HM-RFs). HMRFs are...

Machine learning methods for predicting failures in hard drives: A multiple-instance application (2005)

Joseph F. Murray, Gordon F. Hughes, Dale Schuurmans

We compare machine learning methods applied to a difficult real-world problem: predicting computer hard-drive failure using attributes monitored internally by individual drives. The problem is one of...

Variational bayesian image modelling (2005)

Li Cheng, Feng Jiao, Dale Schuurmans, Shaojun Wang

We present a variational Bayesian framework for performing inference, density estimation and model selection in a special class of graphical models—Hidden Markov Random Fields (HM-RFs). HMRFs are...

Bayesian Sparse Sampling for On-line Reward Optimization (2005)

Tao Wang, Daniel Lizotte, Michael Bowling, Dale Schuurmans

We present an efficient “sparse sampling ” technique for approximating Bayes optimal decision making in reinforcement learning, addressing the well known exploration versus exploitation tradeoff....

Transformation-invariant embedding for image analysis (2004)

Ali Ghodsi, Jiayuan Huang, Dale Schuurmans

Dynamic Web Log Session Identification with Statistical Language Models (2004)

Xiangji Huang, Fuchun Peng, Aijun An, Dale Schuurmans

We present a novel session identification method based on statistical language modeling.

Language and task independent text categorization with simple language models (2003)

Fuchun Peng, Dale Schuurmans, Shaojun Wang

We present a simple method for language independent and task independent text categorization learning, based on character-level n-gram language models. Our approach uses simple information theoretic...

Constraint-based optimization with the minimax decision criterion (2003)

Craig Boutilier, Relu Patrascu, Pascal Poupart, Dale Schuurmans

Abstract. In many situations, a set of hard constraints encodes the feasible configurations of some system or product over which users have preferences. We consider the problem of computing a best...

Latent maximum entropy approach for semantic n-gram language modeling (2003)

Shaojun Wang, Dale Schuurmans, Fuchun Peng

In this paper, we describe a unied probabilistic framework for statistical language modeling|the latent maximum entropy principle|which can eectively incorporate various aspects of natural language,...

Semantic n-gram language modeling with the latent maximum entropy principle (2003)

Shaojun Wang, Dale Schuurmans, Fuchun Peng, Yunxin Zhao

In this paper, we describe a unied probabilistic framework for statistical language modeling|the latent maximum entropy principle|which can eectively incorporate various aspects of natural language,...

Augmenting Naive Bayes Classifiers with Statistical Language Models (2003)

Fuchun Peng, Dale Schuurmans, Shaojun Wang

We augment naive Bayes models with statistical n-gram language models to address shortcomings of the standard naive Bayes text classifier. The result is a generalized naive Bayes classifier

Constraint-based optimization with the minimax decision criterion (2003)

Craig Boutilier, Relu Patrascu, Pascal Poupart, Dale Schuurmans

Abstract. In many situations, a set of hard constraints encodes the feasible con-figurations of some system or product over which users have preferences. We consider the problem of computing a best...

Latent maximum entropy approach for semantic n-gram language modeling (2003)

Shaojun Wang, Dale Schuurmans, Fuchun Peng

In this paper, we describe a unied probabilistic framework for statistical language modeling|the latent maximum entropy principle|which can eectively incorporate various aspects of natural language,...

Language independent authorship attribution using character level language models (2003)

Fuchun Peng, Dale Schuurmans, Viado Keselj, Shaojun Wang

We present a method for computerassisted authorship attribution based on character-level-gram language models. Our approach is based on simple information theoretic principles, and achieves improved...

Combining naive bayes and n-gram language models for text classification (2003)

Language and Task Independent Text Categorization (2003)

Abstract. We augment the naive Bayes model with an n-gram language model to address two shortcomings of naive Bayes text classiers. The chain augmented naive Bayes classiers we propose have two...

With Simple Language, Fuchun Peng, Dale Schuurmans, Shaojun Wang

We present a simple method for language independent and task independent text categorization learning, based on character-level n-gram language models. Our approach uses simple information theoretic...

Combining Naive Bayes and n-Gram Language Models for Text Classification (2003)

Learning Mixture Models with the Latent Maximum Entropy Principle (2003)

We augment the naive Bayes model with an n-gram language model to address two shortcomings of naive Bayes text classifiers.

Shaojun Wang, Dale Schuurmans, Fuchun Peng, Yunxin Zhao

We present a new approach to estimating mixture models based on a new inference principle we have proposed: the latent maximum entropy principle (LME). LME is different both from Jaynes' maximum...

Learning Mixture Models with the Latent Maximum Entropy Principal (2003)

Shaojun Wang, Dale Schuurmans, Fuchun Peng

We present a new approach to estimating mixture models based on a new inference principle we have proposed: the latent maximum entropy principle (LME). LME is different both from Jaynes ’ maximum...

Language independent authorship attribution using character level language models (2003)

Fuchun Peng, Dale Schuurmans, Vlado Keselj, Shaojun Wang

We present a method for computerassisted authorship attribution based on character-level ¤-gram language models. Our approach is based on simple information theoretic principles, and achieves...

Mstream DDoS tool, http://staff.washington.edu/dittrich/misc/mstream.analysis.t xt (2002)

Fuchun Peng, Xiangji Huang, Dale Schuurmans, Nick Cercone

It is commonly believed that word segmentation accuracy is monotonically related to retrieval performance in Chinese information retrieval. In this paper we show that, for Chinese, the relationship...

The latent maximum entropy principle (2002)

Shaojun Wang, Dale Schuurmans, Yunxin Zhao

We present an extension to Jaynes ' maximum entropy principle that handles latent variables. The principle of latent maximum entropy we propose is di#erent from both Jaynes ' maximum...

R-max - a general polynomial time algorithm for near-optimal reinforcement learning (2002)

Ronen I. Brafman, Moshe Tennenholtz, Dale Schuurmans

R-max is a very simple model-based reinforcement learning algorithm which can attain near-optimal average reward in polynomial time. In R-max, the agent always maintains a complete, but possibly...

Investigating the Maximum Likelihood alternative to TD(λ (2002)

Fletcher Lu, Relu Patrascu, Dale Schuurmans

The study of value estimation in Markov reward processes has been dominated by research on temporal dierence methods since the introduction of TD(0) in 1988. Temporal dierence methods are often...

Data perturbation for escaping local maxima in learning (2002)

Gal Elidan, Matan Ninio, Nir Friedman, Dale Schuurmans

Almost all machine learning algorithms---be they for regression, classification or density estimation---seek hypotheses that optimize a score on training data. In most interesting cases, however,...

Algorithm-directed exploration for model-based reinforcement learning in factored MDPs (2002)

Carlos Guestrin, Relu Patrascu, Dale Schuurmans

One of the central challenges in reinforcement learning is to balance the exploration/exploitation tradeoff while scaling up to large problems. Although model-based reinforcement learning has been...

Investigating the Maximum Likelihood alternative to TD(λ (2002)

Fletcher Lu, Relu Patrascu, Dale Schuurmans

The study of value estimation in Markov reward processes has been dominated by research on temporal dierence methods since the introduction of TD(0) in 1988. Temporal dierence methods are often...

Metric-based methods for adaptive model selection and regularization (2002)

R-max - a general polynomial time algorithm for near-optimal reinforcement learning (2002)

and regularization

Ronen I. Brafman, Moshe Tennenholtz, Dale Schuurmans

R-max is a very simple model-based reinforcement learning algorithm which can attain near-optimal average reward in polynomial time. In R-max, the agent always maintains a complete, but possibly...

Investigating the Relationship between Word Segmentation (2002)

Performance And Retrieval, Fuchun Peng, Xiangji Huang, Dale Schuurmans, Nick Cercone

It is commonly believed that word segmentation accuracy is monotonically related to retrieval performance in Chinese information retrieval. In this paper we show that, for Chinese, the relationship...

Applying Machine Learning to Text Segmentation for Information Retrieval (2002)

Xiangji Huang, Fuchun Peng, Dale Schuurmans, Nick Cercone, Stephen Robertson

We propose a self-supervised word segmentation technique for text segmentation in Chinese information retrieval. This method combines the advantages of traditional dictionary based, character based...

On the Existence of Linear Weak Learners and Applications to Boosting (2002)

Shie Mannor, Ron Meir, Yoshua Bengio, Dale Schuurmans

We consider the existence of a linear weak learner for boosting algorithms. A weak learner for binary classification problems is required to achieve a weighted empirical error on the training set...

Using Self-supervised Word Segmentation in Chinese Information Retrieval (2002)

Fuchun Peng, Xiangji Huang, Dale Schuurmans, Nick Cercone, Stephen Robertson

Algorithm-directed exploration for model-based reinforcement learning in factored MDPs (2002)

Carlos Guestrin, Relu Patrascu, Dale Schuurmans

One of the central challenges in reinforcement learning is to balance the exploration/exploitation tradeoff while scaling up to large problems. Although model-based reinforcement learning has been...

Direct value-approximation for factored mdps (2001)

Direct value-approximation for factored mdps (2001)

We present a simple approach for computing reasonable policies for factored Markov decision processes (MDPs), when the optimal value function can be approximated by a compact linear form. Our method...

The exponentiated subgradient algorithm for heuristic boolean programming (2001)

We present a simple approach for computing near-optimal policies in factored Markov decision processes (MDPs), when the optimal value function can be approximated by a compact linear form. Our method...

The exponentiated subgradient algorithm for heuristic boolean programming (2001)

Boolean linear programs (BLPs) are ubiquitous in AI. Satisfiability testing, planning with resource constraints, and winner determination in combinatorial auctions are all examples of this type of...

The exponentiated subgradient algorithm for heuristic boolean programming (2001)

Boolean linear programs (BLPs) are ubiquitous in AI. Satisfiability testing, planning with resource constraints, and winner determination in combinatorial auctions are all examples of this type of...

The exponentiated subgradient algorithm for heuristic boolean programming (2001)

Boolean linear programs (BLPs) are ubiquitous in AI. Satisfiability testing, planning with resource constraints, and winner determination in combinatorial auctions are all examples of this type of...

A Hierarchical EM Approach to Word Segmentation (2001)

Boolean linear programs (BLPs) are ubiquitous in AI. Satisfiability testing, planning with resource constraints, and winner determination in combinatorial auctions are all examples of this type of...

A Hierarchical EM Approach to Word Segmentation (2001)

We propose a simple two-level hierarchical probability model for unsupervised word segmentation. By treating words as strings composed of morphemes /phonemes which are themselves composed of...

Direct value-approximation for factored mdps (2001)

We propose a simple two-level hierarchical probability model for unsupervised word segmentation. By treating words as strings composed of morphemes /phonemes which are themselves composed of...

The exponentiated subgradient algorithm for heuristic boolean programming (2001)

We present a simple approach for computing reasonable policies for factored Markov decision processes (MDPs), when the optimal value function can be approximated by a compact linear form. Our method...

Direct value-approximation for factored mdps (2001)

Boolean linear programs (BLPs) are ubiquitous in AI. Satisfiability testing, planning with resource constraints, and winner determination in combinatorial auctions are all examples of this type of...

Self-supervised Chinese Word Segmentation (2001)

We present a simple approach for computing near-optimal policies in factored Markov decision processes (MDPs), when the optimal value function can be approximated by a compact linear form. Our method...

Metric-Based Methods for Adaptive Model Selection and Regularization (2001)

Abstract. We propose a new unsupervised training method for acquiring probability models that accurately segment Chinese character sequences into words. By constructing a core lexicon to guide...

Using Self-Supervised Word Segmentation in Chinese Information Retrieval (2001)

We present a general approach to model selection and regularization that exploits unlabeled data to adaptively control hypothesis complexity in supervised learning tasks. The idea is to impose a...

Fuchun Peng, Xiangji Huang, Dale Schuurmans, Nick Cercone, Stephen Robertson

We propose a self-supervised word-segmentation technique for Chinese information retrieval. This method combines the advantages of traditional dictionary based approaches with character based...

The exponentiated subgradient algorithm for heuristic boolean programming (2001)

The exponentiated subgradient algorithm for heuristic boolean programming (2001)

Boolean linear programs (BLPs) are ubiquitous in AI. Satisfiability testing, planning with resource constraints, and winner determination in combinatorial auctions are all examples of this type of...

Local search characteristics of incomplete SAT procedures (2000)

Boolean linear programs (BLPs) are ubiquitous in AI. Satisfiability testing, planning with resource constraints, and winner determination in combinatorial auctions are all examples of this type of...

Local search characteristics of incomplete SAT procedures (2000)

Effective local search methods for finding satisfying assignments of CNF formulae exhibit several systematic characteristics in their search. We identify a series of measurable characteristics of...

Local search characteristics of incomplete SAT procedures (2000)

Eective local search methods for nding satisfying assignments of CNF formulae exhibit several systematic characteristics in their search. We identify a series of measurable characteristics of local...

Local search characteristics of incomplete SAT procedures (2000)

Eective local search methods for nding satisfying assignments of CNF formulae exhibit several systematic characteristics in their search. We identify a series of measurable characteristics of local...

Advances in Large Margin Classifiers (2000)

Eective local search methods for nding satisfying assignments of CNF formulae exhibit several systematic characteristics in their search. We identify a series of measurable characteristics of local...

Alexander J. Smola, Alex J. Smola, Peter Bartlett, Dale Schuurmans (Eds.), Peter Bartlett, Bernhard Schölkopf, ...

Contents Preface vii 1 Introduction to Large Margin Classifiers 1 Alex J. Smola, Peter Bartlett, Bernhard Scholkopf, and Dale Schuurmans 2 Large Margin Rank Boundaries for Ordinal Regression 29 Ralf...

An Adaptive Regularization Criterion for Supervised Learning (2000)

Local search characteristics of incomplete SAT procedures (2000)

We introduce a new regularization criterion that exploits unlabeled data to adaptively control hypothesis-complexity in general supervised learning tasks. The technique is based on an abstract...

Monte Carlo inference via greedy importance sampling (2000)

Effective local search methods for finding satisfying assignments of CNF formulae exhibit several systematic characteristics in their search. We identify a series of measurable characteristics of...

Advances in Large Margin Classifiers (2000)

We present a new method for conducting Monte Carlo inference in graphical models which combines explicit search with generalized importance sampling. The idea is to reduce the variance of importance...

Alexander J. Smola, Alex J. Smola, Peter Bartlett, Peter Bartlett, Bernhard Scholkopf, Bernhard Scholkopf, ...

this article also provide a website to obtain the data

Advances in Large Margin Classifiers (2000)

Alexander J. Smola, Alex J. Smola, Peter Bartlett, Peter Bartlett, Bernhard Scholkopf, Bernhard Scholkopf, ...

this article also provide a website to obtain the data

Regularized Greedy Importance Sampling (1999)

Finnegan Southey, Dale Schuurmans, Ali Ghodsi

Greedy importance sampling is an unbiased estimation technique that reduces the variance of standard importance sampling by explicitly searching for modes in the estimation objective. Previous work...

Greedy Importance Sampling (1999)

Efficient Exploration for Optimizing Immediate Reward (1999)

I present a simple variation of importance sampling that explicitly searches for important regions in the target distribution. I prove that the technique yields unbiased estimates, and show...

Advances in Large Margin Classifiers (1999)

We consider the problem of learning an effective behavior strategy from reward. Although much studied, the issue of how to use prior knowledge to scale optimal behavior learning up to real-world...

Alexander J. Smola, Alex J. Smola, Peter Bartlett, Dale Schuurmans (Eds.), Peter Bartlett, Bernhard Schölkopf, ...

this paper are taken from (Herbrich et al., 1999) Smola, Bartlett, Scholkopf, and Schuurmans: Advances in Large Margin Classifiers 1999/03/31 11:08

Greedy importance sampling (1999)

Finnegan Southey, Dale Schuurmans, Ali Ghodsi

Greedy importance sampling is an unbiased estimation technique that reduces the variance of standard importance sampling by explicitly searching for modes in the estimation objective. Previous work...

Greedy importance sampling (1999)

Finnegan Southey, Dale Schuurmans, Ali Ghodsi

Greedy importance sampling is an unbiased estimation technique that reduces the variance of standard importance sampling by explicitly searching for modes in the estimation objective. Previous work...

Greedy importance sampling (1999)

Finnegan Southey, Dale Schuurmans, Ali Ghodsischool, Computer Science

Abstract Greedy importance sampling is an unbiased estimation technique that re-duces the variance of standard importance sampling by explicitly searching for modes in the estimation objective....

Boosting in the limit: Maximizing the margin of learned ensembles (1998)

Adam J. Grove, Dale Schuurmans

The "minimum margin" of an ensemble classifier on a given training set is, roughly speaking, the smallest vote it gives to any correct training label. Recent work has shown that the...

Boosting in the limit: Maximizing the margin of learned ensembles (1998)

Adam Grove, Dale Schuurmans

The "minimum margin" of an ensemble classifier on a given training set is, roughly speaking, the smallest vote it gives to any correct training label. Recent work has shown that the...

Boosting in the limit: Maximizing the margin of learned ensembles (1998)

Adam J. Grove, Dale Schuurmans

The "minimum margin" of an ensemble classifier on a given training set is, roughly speaking, the smallest vote it gives to any correct training label. Recent work has shown that the...

General convergence results for linear discriminant updates (1997)

On Learning Hierarchical Classifications (1997)

Abstract. The problem of learning linear-discriminant concepts can be solved by various mistake-driven update procedures, including the Winnow family of algorithms and the well-known Perceptron...

Russell Greiner, Adam Grove, Dale Schuurmans

Many significant real-world classification tasks involve a large number of categories which are arranged in a hierarchical structure; for example, classifying documents into subject categories under...

General Convergence Results for Linear Discriminant Updates (1997)

Characterizing the Generalization Performance of Model Selection Strategies (1997)

The problem of learning linear-discriminant concepts can be solved by various mistakedriven update procedures, including the Winnow family of algorithms and the wellknown Perceptron algorithm. In...

Dale Schuurmans, Lyle H. Ungar, Dean P. Foster

: We investigate the structure of model selection problems via the bias/variance decomposition. In particular, we characterize the essential structure of a model selection task by the bias and...

General Convergence Results for Linear Discriminant Updates (1997)

Learning Bayesian Nets that Perform Well (1997)

The problem of learning linear-discriminant concepts can be solved by various mistakedriven update procedures, including the Winnow family of algorithms and the wellknown Perceptron algorithm. In...

Russell Greiner, Adam J. Grove, Dale Schuurmans

A Bayesian net (BN) is more than a succinct way to encode a probabilistic distribution; it also corresponds to a function used to answer queries. A BN can therefore be evaluated by the accuracy of...

Fast (Distribution Specific) Learning (1997)

General Convergence Results for Linear Discriminant Updates (1997)

PAC-learning results are often criticized for demanding impractically large training samples. The common wisdom is that these large samples follow from the worst case nature of the analysis, and...

A New Metric-Based Approach to Model Selection (1997)

The problem of learning linear discriminant concepts can be solved by various mistake-driven update procedures, including the Winnow family of algorithms and the well-known Perceptron algorithm. In...

Characterizing the Generalization Performance of Model Selection Strategies (1997)

We introduce a new approach to model selection that performs better than the standard complexitypenalization and hold-out error estimation techniques in many cases. The basic idea is to exploit the...

Dale Schuurmans, Lyle H. Ungar, Dean P. Foster

We investigate the structure of model selection problems via the bias/variance decomposition. In particular, we characterize the essential aspects of a model selection task by the bias and variance...

Learning Bayesian Nets that Perform Well (1997)

Russell Greiner, Adam J. Grove, Dale Schuurmans

A Bayesian net (BN) is more than a succinct way to encode a probabilistic distribution; it also corresponds to a function used to answer queries. A BN can therefore be evaluated by the accuracy of...

D.: On learning hierarchical Classifications (1997)

Russell Greiner, Adam Grove, Dale Schuurmans

Many significant real-world classification tasks involve a large number of categories which are arranged in a hierarchical structure; for example, classifying documents into subject categories under...

Characterizing the generalization performance of model selection strategies (1997)

Effective classification learning [microform] / (1996)

We investigate the structure of model selection problems via the bias/variance decomposition. In particular, we characterize the essential aspects of a model selection task by the bias and variance...

Schuurmans, Dale.

Thesis (Ph. D.)--University of Toronto, 1996.

Sequential PAC Learning (1995)

Sequential PAC Learning (1995)

We consider the use of "on-line" stopping rules to reduce the number of training examples needed to pac-learn. Rather than collect a large training sample that can be proved sufficient to...

Characterizing Rational versus Exponential Learning Curves (1995)

We consider the use of "on-line" stopping rules to reduce the number of training examples needed to pac-learn. Rather than collect a large training sample that can be proved sufficient to...

Practical PAC Learning (1995)

. We consider the standard problem of learning a concept from random examples. Here a learning curve can be defined to be the expected error of a learner's hypotheses as a function of training...

Practical PAC Learning (1995)

We present new strategies for "probably approximately correct" (pac) learning that use fewer training examples than previous approaches. The idea is to observe training examples...

Probabilistic hill-climbing (1994)

We present new strategies for "probably approximately correct" (pac) learning that use fewer training examples than previous approaches. The idea is to observe training examples...

William W. Cohen, Russell Greiner, Dale Schuurmans

Abstract: Many learning tasks involve searching through a discrete space of performance elements, seeking an element whose future utility is expected to be high. As the task of finding the global...

Learning an Optimally Accurate Representation System (1994)

Russell Greiner, Dale Schuurmans

. A default theory can sanction different, mutually incompatible, answers to certain queries. We can identify each such theory with a set of related credulous theories, each of which produces but a...

Learning Default Concepts (1994)

Learning Default Concepts (1994)

Classical concepts, based on necessary and sufficient defining conditions, cannot classify logically insufficient object descriptions. Many reasoning systems avoid this limitation by using...

Learning to Classify Incomplete Examples (1993)

Classical concepts, based on necessary and sufficient defining conditions, cannot classify logically insufficient object descriptions. Many reasoning systems avoid this limitation by using...