Predicting homologous signaling pathways using machine learning (2009)
Bostan, Babak, Greiner, Russell, Szafron, Duane, Lu, Paul
Motivation: In general, each cell signaling pathway involves many proteins, each with one or more specific roles. As they are essential components of cell activity, it is important to understand how...
BIOINFORMATICS Using Rank-1 Biclusters to Classify Microarray Data (2008)
Nasimeh Asgarian, Russell Greiner
Motivation: A DNA-microarray measures the gene expression levels of tens of thousands of genes for a particular sample, corresponding to some specific experimental condition. Our goal is to learn a...
On Pearcy, Paul Lu, Duane Szafron, Russell Greiner, David S. Wishart, Alona Fyshe, ...
•Drosophila melanogaster •Homo sapiens •Mus musculus •Plasmodium falciparum
Tim Van Allen, Ajit Singh, Russell Greiner, Peter Hooper
A Bayesian belief network models a joint distribution over variables using a DAG to represent variable dependencies and network parameters to represent the conditional probability of each variable...
Suffix Tree Optimized Methods of Prediction Suffix trees 2: (2008)
Bioinformatics I, Brett Poulin, Duane Szafron, Russell Greiner, Paul Lu, Roman Eisner, ...
www.cs.ualberta.ca/~bioinfo What’s the goal? Our ultimate goal is high-throughput protein function prediction from amino acid sequences. Our current system, Proteome Analyst (PA) 1 is a strong step...
PREDICTING PROTEIN FUNCTION USING MACHINE-LEARNED HIERARCHICAL CLASSIFIERS (2008)
Roman Eisner, Duane Szafron, Paul Lu, Warren Gallin, Russell Greiner
Trying to determine the structure of a protein by UV spectroscopy was like trying to determine the structure of a piano by listening to the sound it made while being dropped down a flight of stairs.
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...
Appendix to “Structural Extension to Logistic Regression” (2008)
Russell Greiner, Xiaoyuan Su, Bin Shen, Wei Zhou
This report contains material that complements the article (GSSZ05). Appendix
Learning a Classification-based Glioma Growth Model Using MRI Data (2008)
Marianne Morris, Russell Greiner, Jörg S, Albert Murtha, Mark Schmidt
Abstract — Gliomas are malignant brain tumors that grow by invading adjacent tissue. We propose and evaluate a 3D classification-based growth model, CDM, that predicts how a glioma will grow at a...
Russell Greiner, Joseph Likuski
"Explanation-based learning " — i.e., incorporating new redundant rules suggested by earlier problem solving experiences — is an attempt to speed up problem solving....
See also EXPLANATION; KNOWLEDGE REPRESENTATION; PROBLEM SOLVING (2008)
in future problems. Also, although our description assumes the background theory to be “perfect, ” there have been extensions to deal with theories that are incomplete, intractable, or...
To Appear in Artificial Intelligence Journal Learning Cost-Sensitive Active Classifiers ∗ (2008)
Russell Greiner, Dan Roth, Adam J. Grove
Most classification algorithms are “passive”, in that they assign a class label to each instance based only on the description given, even if that description is incomplete. By contrast, an...
Learning a Dynamic Classification Method to Detect Faces and Identify Facial Expression (2008)
Ramana Isukapalli, Russell Greiner, Ahmed Elgammal
Abstract. While there has been a great deal of research in face detection and recognition, there has been very limited work on identifying the expression on a face. Many current face detection...
Learning to Detect Objects of Many Classes Using Binary Classifiers (2008)
Ramana Isukapalli, Ahmed Elgammal, Russell Greiner
Abstract. Viola and Jones [VJ] demonstrate that cascade classification methods can successfully detect objects belonging to a single class, such as faces. Detecting and identifying objects that...
Learning Robust Object Recognition Strategies (2008)
Ilya Levner, Vadim Bulitko, Lihong Li, Greg Lee, Russell Greiner
Automated image interpretation is an important task in numerous applications ranging from security systems to natural resource inventorization based on remote-sensing. Recently, a second generation...
Heterogeneous Stacking for Classification-Driven Watershed Segmentation (2008)
Ilya Levner, Hong Zhang, Russell Greiner
Marker-driven watershed segmentation attempts to extract seeds that indicate the presence of objects within an image. These markers are subsequently used to enforce regional minima within a...
Heterogeneous Stacking for Classification-Driven Watershed Segmentation (2008)
Ilya Levner, Hong Zhang, Russell Greiner
Marker-driven watershed segmentation attempts to extract seeds that indicate the presence of objects within an image. These markers are subsequently used to enforce regional minima within a...
Speeding up planning in Markov decision processes via automatically constructed abstraction (2008)
Ro Isaza, Csaba Szepesvári, Vadim Bulitko, Russell Greiner
In this paper, we consider planning in stochastic shortest path (SSP) problems, a subclass of Markov Decision Problems (MDP). We focus on medium-size problems whose state space can be fully...
t standard learning algorithms are rather limited and fragile. Many of my results extend these algorithms, and analyses, to produce more robust and more effective learning systems. In the last few...
How an Expert can use Imperfect Knowledge to Improve an Imperfect Theory (2007)
Russell Greiner, Jie Cheng, Christian Darken
This report addresses the challenge of using auxiliary information I A to improve a given theory, encoded as a belief net BE . In contrast with many other "knowledge revision" systems, we...
In Er, Yan Xiao, Russell Greiner
This paper describes the architecture of an efficient plan verifier that can first detect faults in a planner's plans and use these observed errors to identify possible problems in the...
Exploiting the Absence of Irrelevant Information: What You Don't Know (2007)
Help You, R. Bharat Rao, Russell Greiner, Thomas Hancock
Most inductive inference algorithms are designed to work most effectively when their training data contain completely specified labeled samples. In many environments, however, the person collecting...
A Distributed Plan Verifier (2007)
This paper describes the architecture of an efficient plan verifier that can first detect faults in a planner's plans and use these observed errors to identify possible problems in the...
The AIxploratorium: A vision for AI and the web (2007)
Proceedings of the \Eective Interactive AI Resources " (IJCAI'01 workshop) Seattle, Aug 2001. The web is making fundamental changes to the eld of articial intelligence (AI), ranging...
Russell Greiner, Ryan Hayward, Michael Molloy
A probabilistic boolean expression (PBE) consists of a boolean expression over a set of boolean variables, each with a corresponding cost and probability value that indicates respectively the cost of...
Budgeted Learning, Part II: The Naïve-Bayes Case (2007)
Daniel J. Lizotte, Omid Madani, Russell Greiner
There is almost always a cost associated with acquiring training data. We consider the situation where the learner, with a xed budget, may `purchase' data during training. In particular, we...
Learning Bayesian Networks from Data: An Information-Theory Based Approach (2007)
Jie Cheng, Russell Greiner, Jonathan Kelly, David Bell, Weiru Liu
This paper provides algorithms that use an information-theoretic analysis to learn Bayesian network structures from data. Based on our three-phase learning framework, we develop efficient algorithms...
Mind Change Optimal Learning of Bayes Net Structure". O.Schulte (2007)
Oliver Schulte, Russell Greiner
Abstract. This paper analyzes the problem of learning the structure of a Bayes net (BN) in the theoretical framework of Gold’s learning paradigm. Bayes nets are one of the most prominent formalisms...
Mind Change Optimal Learning of Bayes Net Structure". O.Schulte (2007)
Oliver Schulte, Wei Luo, Russell Greiner
Abstract. This paper analyzes the problem of learning the structure of a Bayes net (BN) in the theoretical framework of Gold’s learning paradigm. Bayes nets are one of the most prominent formalisms...
The Path-A metabolic pathway prediction web server (2006)
Luca Pireddu, Duane Szafron, Paul Lu, Russell Greiner
Pathway Analyst (Path-A) is a publicly available web server
The Path-A metabolic pathway prediction web server (2006)
Luca Pireddu Duane, Duane Szafron, Paul Lu, Russell Greiner
Pathway Analyst (Path-A) is a publicly available web server (http://path-a.cs.ualberta.ca) that predicts metabolic pathways. It takes a FASTA format file containing a set of query protein sequences...
Efficient spatial classification using decoupled conditional random fields (2006)
Chi-hoon Lee, Russell Greiner, Osmar Zaïane
Abstract. We present a discriminative method to classify data that have interdependencies in 2-D lattice. Although both Markov Random Fields (MRFs) and Conditional Random Fields (CRFs) are well-known...
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...
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...
The Path-A metabolic pathway prediction web server (2006)
Pireddu, Luca, Szafron, Duane, Lu, Paul, Greiner, Russell
Pathway Analyst (Path-A) is a publicly available web server (http://path-a.cs.ualberta.ca) that predicts metabolic pathways. It takes a FASTA format file containing a set of query protein sequences...
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...
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...
Learning and Classifying under Hard Budgets (2005)
Abstract. Since resources for data acquisition are seldom infinite, both learners and classifiers must act intelligently under hard budgets. In this paper, we consider problems in which feature...
Classification-based Glioma Diffusion Modeling (2005)
Marianne Morris, Russell Greiner, Jörg S, Albert Murtha, Mark Schmidt
Abstract. Gliomas are diffuse, invasive brain tumors. We propose a 3D classification-based diffusion model, cdm, that predicts how a glioma will grow at a voxel-level, on the basis of features...
Support vector random fields for spatial classification (2005)
Chi-hoon Lee, Russell Greiner, Mark Schmidt
Abstract. In this paper we propose Support Vector Random Fields (SVRFs), an extension of Support Vector Machines (SVMs) that explicitly models spatial correlations in multi-dimensional data. SVRFs...
Learning and Classifying under Hard Budgets (2005)
Abstract. Since resources for data acquisition are seldom infinite, both learners and classifiers must act intelligently under hard budgets. In this paper, we consider problems in which feature...
Paul Lu, Duane Szafron, Russell Greiner, David S. Wishart, Alona Fyshe, Brett Poulin, ...
PA-GOSUB (Proteome Analyst: GO Molecular Function and Subcellular Localization) is a publiclyavailable, web-based, searchable, and downloadable database that contains the sequences, predicted GO...
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...
Budgeted learning of bounded active classifiers (2005)
Abstract. Since resources for data acquisition are seldom infinite, the need exists for learners and classifiers that act intelligently under hard budgets. In this paper, we consider problems in...
Classification-based Glioma Diffusion Modeling (2005)
Marianne Morris, Russell Greiner, Jörg S, Albert Murtha, Mark Schmidt
Abstract. Gliomas are diffuse, invasive brain tumors. We propose a 3D classification-based diffusion model, cdm, that predicts how a glioma will grow at a voxel-level, on the basis of features...
Segmenting brain tumor with conditional random fields and support vector machines (2005)
Chi-hoon Lee, Mark Schmidt, Albert Murtha, Aalo Bistritz, Jöerg S, Russell Greiner
Abstract. Markov Random Fields (MRFs) are a popular and wellmotivated model for many medical image processing tasks such as segmentation. Discriminative Random Fields (DRFs), a discriminative...
Russell Greiner, Ryan Hayward, Magdalena Jankowska, Michael Molloy
Many tasks require evaluating a specified Boolean expression ϕ over a set of probabilistic tests whose costs and success probabilities are each known. A strategy specifies when to perform which...
Support vector random fields for spatial classification (2005)
Chi-hoon Lee, Russell Greiner, Mark Schmidt
Abstract. In this paper we propose Support Vector Random Fields (SVRFs), an extension of Support Vector Machines (SVMs) that explicitly models spatial correlations in multi-dimensional data. SVRFs...
Lu, Paul, Szafron, Duane, Greiner, Russell, Wishart, David S., Fyshe, Alona, Pearcy, Brandon, ...
PA-GOSUB (Proteome Analyst: Gene Ontology Molecular Function and Subcellular Localization) is a publicly available, web-based, searchable and downloadable database that contains the sequences,...
Focus of attention in reinforcement learning (2004)
Lihong Li, Vadim Bulitko, Russell Greiner
Abstract: One key topic in reinforcement learning is function approximation which is critical for successfully applying reinforcement learning to domains with large state spaces. Unfortunately,...
Duane Szafron, Paul Lu, Russell Greiner, David S. Wishart, Brett Poulin, Roman Eisner, ...
explanations in a web-based tool for high-throughput proteome annotations
Batch reinforcement learning with state importance (2004)
Lihong Li, Vadim Bulitko, Russell Greiner
Abstract. We investigate the problem of using function approximation in reinforcement learning where the agent’s policy is represented as a classifier mapping states to actions. High classification...
Paul Lu, Duane Szafron, Russell Greiner, David S. Wishart, Alona Fyshe, On Pearcy, ...
a searchable database of model organism
Focus of attention in reinforcement learning (2004)
Lihong Li, Vadim Bulitko, Russell Greiner
Abstract: Classification-based reinforcement learning (RL) methods have recently been proposed as an alternative to the traditional value-function based methods. These methods use a classifier to...
Duane Szafron, Paul Lu, Russell Greiner, David S. Wishart, Brett Poulin, Roman Eisner, ...
Proteome Analyst (PA) (http://www.cs.ualberta.ca/ ~bioinfo/PA/) is a publicly-available, high-throughput, Web-based system for predicting various properties of each protein in an entire proteome....
Focus of attention in sequential decision making (2004)
Lihong Li, Vadim Bulitko, Russell Greiner
We investigate the problem of using function approximation in reinforcement learning (RL) where the agent’s control policy is represented as a classifier mapping states to actions. The innovation...
Hierarchical probabilistic relational models for collaborative filtering (2004)
This paper applies Probabilistic Relational Models (PRMs) to the Collaborative Filtering task, focussing on the EachMovie data set. We first learn a standard PRM, and show that its performance is...
The budgeted multi-armed bandit problem (2004)
Omid Madani, Daniel J. Lizotte, Russell Greiner
The following coins problem is a version of a multi-armed bandit problem where one has to select from among a set of objects, say classifiers, after an experimentation phase that is constrained by a...
Duane Szafron, Paul Lu, Russell Greiner, David Wishart, Brett Poulin, Roman Eisner, ...
Proteome Analyst (PA)
Batch reinforcement learning with state importance (2004)
Lihong Li, Vadim Bulitko, Russell Greiner
Abstract. We investigate the problem of using function approximation in reinforcement learning where the agent’s policy is represented as a classifier mapping states to actions. High classification...
Szafron, Duane, Lu, Paul, Greiner, Russell, Wishart, David S., Poulin, Brett, Eisner, Roman, ...
Proteome Analyst (PA) (http://www.cs.ualberta.ca/~bioinfo/PA/) is a publicly available, high-throughput, web-based system for predicting various properties of each protein in an entire proteome....
Ajit Paul Singh, Wolfgang Pauli, Ajit Paul Singh, Russell Greiner, Peter Hooper, Terry Caelli
of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. The author reserves all other publication and other rights in association with the...
Duane Szafron, Paul Lu, Russell Greiner, David Wishart, Zhiyong Lu, Brett Poulin, ...
Modern sequencing technology permits sequencing of entire genomes, whose gene sequences require annotation. It is too time consuming to predict the properties of each protein sequence manually and to...
Explaining Naive Bayes Classifications (2003)
Duane Szafron, Russell Greiner, Paul Lu, David Wishart, Cam Macdonell, John Anvik, ...
Naïve Bayes classifiers, a popular tool for predicting the labels of query instances, are typically learned from a training set. However, since many training sets contain noisy data, a classifier...
Automated feature extraction for object recognition (2003)
Ilya Levner, Vadim Bulitko, Lihong Li, Greg Lee, Russell Greiner
Automated image interpretation is an important task in numerous applications ranging from security systems to natural resource inventorization based on remote-sensing. Recently, a second generation...
Improving an adaptive image interpretation system by leveraging (2003)
Lihong Li, Vadim Bulitko, Russell Greiner, Ilya Levner
Abstract Automated image interpretation is an important task innumerous applications ranging from security systems to natural resource inventorization based on remote-sensing.Recently, a second...
Budgeted Learning, Part I: The Multi-Armed Bandit Case (2003)
Omid Madani Daniel, Daniel J. Lizotte, Russell Greiner
We introduce and motivate the task of learning under a budget. We focus on a basic problem in this space: selecting the optimal bandit after a period of experimentation in a multi-armed bandit...
Improving an adaptive image interpretation system by leveraging (2003)
Lihong Li, Vadim Bulitko, Russell Greiner, Ilya Levner
Automated image interpretation is an important task in numerous applications ranging from security systems to natural resource inventorization based on remote-sensing. Recently, a second generation...
Adaptive image interpretation : A spectrum of machine learning problems (2003)
Vadim Bulitko, Lihong Li, Greg Lee, Russell Greiner, Ilya Levner
Automated image interpretation is an important task in numerous applications ranging from security systems to natural resource inventorization based on remote-sensing. Recently, a second generation...
Towards automated creation of image interpretation systems (2003)
Ilya Levner, Vadim Bulitko, Lihong Li, Greg Lee, Russell Greiner
Abstract. Automated image interpretation is an important task in numerous applications ranging from security systems to natural resource inventorization based on remote-sensing. Recently, a second...
Use of Off-line Dynamic Programming for Efficient Image Interpretation (2003)
Ramana Isukapalli Crawfords, Ramana Isukapalli, Russell Greiner
An interpretation system finds the likely mappings from portions of an image to real-world objects. An interpretation policy specifies when to apply which imaging operator, to which portion of the...
Discriminative parameter learning of general Bayesian network classifiers (2003)
Bin Shen, Petr Musilek, Xiaoyuan Su, Russell Greiner, Corrine Cheng
Greiner and Zhou [1] presented ELR, a discriminative parameter-learning algorithm that maximizes conditional likelihood (CL) for a fixed Bayesian Belief Network (BN) structure, and demonstrated that...
RLL-1: A Representation Language Language. Supplement. Details of RLL-1. (2002)
Greiner,Russell, Lenat,Douglas B.
This paper includes many implementation level details about the RLL-1 system, described in a companion paper, RLL-1: A Representation Language Language (Heuristic Programming Project Working Paper...
MRS Manual. Multiple Representation System, (2002)
Genesereth,Michael R., Greiner,Russell, Smith,David E.
MRS is a knowledge representation system intended for use by AI researchers in building expert systems. It offers a diverse repertory of commands for asserting and retrieving information, with...
Russell Greiner, Xiaoyuan Su, Bin Shen, Wei Zhou
Abstract. Bayesian belief nets (BNs) are often used for classification tasks — typically to return the most likely class label for each specified instance. Many BN-learners, however, attempt to...
Russell Greiner, Wei Zhou, Xiaoyuan Su, Bin Shen
Bayesian belief nets (BNs) are often used for classification tasks — typically to return the most likely class label for each specified instance. Many BN-learners, however, attempt to find the BN...
Performance of lookahead control policies in the face of abstractions and approximations (2002)
Omid Madani, Vadim Bulitko, Ilya Levner, Russell Greiner
Abstract. This paper explores the formulation of image interpretation as a Markov Decision Process (MDP) problem, highlighting the important assumptions in the MDP formulation. Furthermore state...
Optimal Depth-First Strategies for And-Or Trees (2002)
Russell Greiner, Ryan Hayward, Michael Molloy
Many tasks require evaluating a specied boolean expression # over a set of probabilistic tests where we know the probability that each test will succeed, and also the cost of performing each test. A...
Russell Greiner, Wei Zhou, Xiaoyuan Su, Bin Shen
Bayesian belief nets (BNs) are often used for classification tasks — typically to return the most likely class label for each specified instance. Many BN-learners, however, attempt to find the BN...
Performance of lookahead control policies in the face of abstractions and approximations (2002)
Ilya Levner, Vadim Bulitko, Omid Madani, Russell Greiner
Abstract. This paper explores the formulation of image interpretation as a Markov Decision Process (MDP) problem, highlighting the important assumptions in the MDP formulation. Furthermore state...
Bayesian belief nets (BNs) are often used for classification tasks — typically to return the most likely “class label ” for each specified instance. Many BN-learners, however, attempt to find...
Optimal depth-first strategies for and-or trees (2002)
Many tasks require evaluating a speci£ed boolean expression ϕ over a set of probabilistic tests where we know the probability that each test will succeed, and also the cost of performing each test....
Optimal depth-first strategies for and-or trees (2002)
Many tasks require evaluating a speci£ed boolean expression ϕ over a set of probabilistic tests where we know the probability that each test will succeed, and also the cost of performing each test....
Optimal Depth-First Strategies for And-Or Trees (2002)
Russell Greiner And, Russell Greiner, Ryan Hayward, Michael Molloy
A probabilistic boolean expression (PBE) consists of a boolean expression over a set of boolean variables, each with a corresponding cost and probability value that indicates respectively the cost of...
Bayesian belief nets (BNs) are often used for classification tasks --- typically to return the most likely "class label" for each specified instance. Many BN-learners, however, attempt to...
Russell Greiner, Xiaoyuan Su, Bin Shen, Wei Zhou
Abstract. Bayesian belief nets (BNs) are often used for classification tasks — typically to return the most likely class label for each specified instance. Many BN-learners, however, attempt to...
1 Abstract Learning Bayesian Networks from Data: An Information-Theory Based Approach (2001)
Jie Cheng, Russell Greiner, Jonathan Kelly, David Bell, Weiru Liu
This paper provides algorithms that use an information-theoretic analysis to learn Bayesian network structures from data. Based on our three-phase learning framework, we develop efficient algorithms...
Learning Bayesian Belief Network Classifiers: Algorithms and System (2001)
This paper investigates the methods for learning predictive classifiers based on Bayesian belief networks (BN)-- primarily unrestricted Bayesian networks and Bayesian multinets. We present our...
Learning Bayesian Belief Network Classifiers: Algorithms and System (2001)
Abstract. This paper investigates the methods for learning predictive classifiers based on Bayesian belief networks (BN) – primarily unrestricted Bayesian networks and Bayesian multi-nets. We...
Efficient interpretation policies (2001)
Ramana Isukapalli, Russell Greiner
Many imaging systems seek a good interpretation of the scene presented--- i.e., a plausible (perhaps optimal) mapping from aspects of the scene to real-world objects. This paper addresses the issue...
Efficient car recognition policies (2001)
Ramana Isukapalli, Russell Greiner
Many tasks require an imaging system to identify an object, such as the type of a car; in many cases, it is critical to make this identification quickly, as well as accurately. This paper addresses...
Recent results have shown that \Bayesian classiers", including NaveBayes, perform extremely well as classiers. Essentially all of the associated learners seek the parameters that maximize...
Bayesian error-bars for belief net inference (2001)
Tim Van Allen, Russell Greiner, Peter Hooper
A Bayesian Belief Network (BN) is a model of a joint distribution over a finite set of variables, with a DAG structure to represent the immediate dependencies between the variables, and a set of...
Learning Bayesian Belief Network Classifiers: Algorithms and System (2001)
This paper investigates the methods for learning Bayesian belief network (BN) based predictive models for classification. Our primary interests are in the unrestricted Bayesian network and Bayesian...
1 Abstract Learning Bayesian Networks from Data: An Information-Theory Based Approach (2001)
Jie Cheng, Russell Greiner, Jonathan Kelly, David Bell, Weiru Liu
This paper provides algorithms that use an information-theoretic analysis to learn Bayesian network structures from data. Based on our three-phase learning framework, we develop efficient algorithms...
Predicting UNIX Command Lines (2000)
Benjamin Korvemaker, Russell Greiner
asing the current probability when a previously encountered stub is observed. This method obtains nearly 75% accuracy when predicting the 5 most likely commands 3 . This was a substantial improvement...
Predicting UNIX Command Lines: Adjusting to User Patterns (2000)
Benjamin Korvemaker, Russell Greiner
As every user has his own idiosyncrasies and preferences, an interface that is honed for one user may be problematic for another. To accommodate a diverse range of users, many computer applications...
Predicting UNIX Command Lines: Adjusting to User Patterns (2000)
Benjamin Korvemaker, Russell Greiner
As every user has his own idiosyncrasies and preferences, an interface that is honed for one user may be problematic for another. To accommodate a diverse range of users, many computer applications...
Comparing Bayesian Network Classifiers (1999)
In this paper, we empirically evaluate algorithms for learning four types of Bayesian network (BN) classifiers -- Naïve-Bayes, tree augmented Naïve-Bayes, BN augmented Naïve-Bayes and general BNs,...
The Trials and Tribulations of Building an Adaptive User Interface (1999)
Benjamin Korvemaker, Russell Greiner
As every user has his own ideosyncracies and preferences, an interface that is honed for one user may be problematic for another. To accomodate a diverse range of users, many computer applications...
The Complexity of Revising Logic Programs (1999)
A rule-based program will return a set of answers to each query. An impure program, which includes the Prolog cut "!" and "not(\Delta)" operators, can return different answers if...
Learning Accurate Belief Nets (1999)
Bayesian belief nets (BNs) are typically used to answer a range of queries, where each answer requires computing the probability of a particular hypothesis given some specified evidence. An effective...
Comparing Bayesian Network Classifiers (1999)
In this paper, we empirically evaluate algorithms for learning four types of Bayesian network (BN) classifiers -- Naïve-Bayes, tree augmented Naïve-Bayes (TANs), BN augmented Naïve-Bayes (BANs)...
Comparing Bayesian Network Classifiers (1999)
In this paper, we empirically evaluate algorithms for learning four Bayesian network (BN) classifiers: Naïve-Bayes, tree augmented Naïve-Bayes (TANs), BN augmented NaïveBayes (BANs) and general...
Russell Greiner, Christian Darken, N. Iwan Santoso
Many tasks require "reasoning" --- i.e., deriving conclusions from a corpus of explicitly stored information --- to solve their range of problems. An ideal reasoning system would produce...
Why experimentation can be better than perfect guidance (1997)
Tobias Scheffer, Russell Greiner, Christian Darken
The full version of this paper appeared at ICML-97. Many problems correspond to the classical control task of determining the appropriate control action to take, given some (sequence of)...
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...
Knowing What Doesn't Matter: Exploiting the Omission of Irrelevant Data (1997)
Russell Greiner, Adam J. Grove, Alexander Kogan
Most learning algorithms work most effectively when their training data contain completely specified labeled samples. In many diagnostic tasks, however, the data will include the values of only some...
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...
Fast (Distribution Specific) Learning (1997)
Dale Schuurmans, Russell Greiner
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...
Why Experimentation can be better than "Perfect Guidance" (1997)
Tobias Scheffer, Russell Greiner, Christian Darken
Many problems correspond to the classical control task of determining the appropriate control action to take, given some (sequence of) observations. One standard approach to learning these control...
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...
Why Experimentation can be better than "Perfect Guidance" (1997)
Tobias Scheffer, Russell Greiner, Christian Darken
Many problems correspond to the classical control task of determining the appropriate control action to take, given some (sequence of) observations. One standard approach to learning these control...
The Relevance of Relevance (1997)
Devika Subramanian, Russell Greiner
Introduction With too little information, reasoning and learning systems cannot work effectively. Too much information can also cause the performance of these systems to degrade, in both accuracy and...
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...
Learning active classifiers (1996)
Russell Greiner, Adam J. Grove, Dan Roth
Most classification algorithms are "passive", in that they assign a class-label to each instance based only on the description given, even if that description is incomplete. By contrast, an...
Learning Active Classifiers (1996)
Russell Greiner, Adam J. Grove, Dan Roth
Many classification algorithms are "passive", in that they assign a class-label to each instance based only on the description given, even if that description is incomplete. In contrast, an...
Exploiting the Omission of Irrelevant Data (1996)
Russell Greiner, Adam J. Grove, Alexander Kogan
Most learning algorithms work most effectively when their training data contain completely specified labeled samples. In many diagnostic tasks, however, the data will include the values of only some...
Learning Active Classifiers (1996)
Russell Greiner, Adam J. Grove, Dan Roth
Many classification algorithms are "passive", in that they assign a class-label to each instance based only on the description given, even if that description is incomplete. In contrast, an...
The Complexity of Revising Logic Programs (1996)
A rule-based program will return a set of answers to each query. An impure program, which includes the Prolog cut "!" and "not(\Delta)" operators, can return different answers if...
PALO: A Probabilistic Hill-Climbing Algorithm (1996)
Many learning systems search through a space of possible performance elements, seeking an element whose expected utility, over the distribution of problems, is high. As the task of finding the...
The complexity of theory revision (1995)
A knowledge-based system uses its database (a.k.a. its "theory") to produce answers to the queries it receives. Unfortunately, these answers may be incorrect if the underlying...
An optimized theory revision module (1995)
Russell Greiner, R. Bharat Rao, Glenn Meredith
Theory revision systems typically use a set of theory-to-theory transformations f k g to hill-climb from a given initial theory to a new theory whose empirical accuracy, over a given set of labeled...
Sequential PAC Learning (1995)
Dale Schuurmans, Russell Greiner
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...
Sequential PAC Learning (1995)
Dale Schuurmans, Russell Greiner
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...
Dale Schuurmans, Russell Greiner
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...
PALO: A Probabilistic Hill-Climbing Algorithm (1995)
Many learning systems search through a space of possible performance elements, seeking an element whose expected utility, over the distribution of problems, is high. As the task of finding the...
Dale Schuurmans, Russell Greiner
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...
The complexity of theory revision (1995)
A knowledge-based system uses its database (a k a its "theory") to produce answers to the queries it receives Unfortunately, these an swere may be incorrect if the underlying theory...
Probabilistic hill-climbing (1994)
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 to Select Useful Landmarks (1994)
Russell Greiner, Ramana Isukapalli
To navigate effectively, an autonomous agent must be able to quickly and accurately determine its current location. Given an initial estimate of its position (perhaps based on dead-reckoning) and an...
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...
Knowing What Doesn't Matter: Exploiting Omitted Superfluous Data (1994)
Russell Greiner, Thomas Hancock, R. Bharat Rao
Most inductive inference algorithms (i.e., "learners") work most effectively when their training data contain completely specified labeled samples. In many diagnostic tasks, however, the...
Learning to Select Useful Landmarks (1994)
Russell Greiner, Ramana Isukapalli
To navigate effectively, an autonomous agent must be able to quickly and accurately determine its current location. Given an initial estimate of its position (perhaps based on dead-reckoning) and an...
Learning Default Concepts (1994)
Dale Schuurmans, Russell Greiner
Classical concepts, based on necessary and sufficient defining conditions, cannot classify logically insufficient object descriptions. Many reasoning systems avoid this limitation by using...
Learning Default Concepts (1994)
Dale Schuurmans, Russell Greiner
Classical concepts, based on necessary and sufficient defining conditions, cannot classify logically insufficient object descriptions. Many reasoning systems avoid this limitation by using...
Theory Revision in Fault Hierarchies (1994)
Pat Langley, George Drastal, R. Bharat Rao, Russell Greiner
The fault hierarchy representation is widely used in expert systems for the diagnosis of complex mechanical devices. On the assumption that an appropriate bias for a knowledge representation language...
Knowing What Doesn't Matter: Exploiting The Omission of Irrelevant Data (1994)
Russell Greiner, Adam J. Grove, Alexander Kogan
Most learning algorithms work most effectively when their training data contain completely specified labeled samples. In many diagnostic tasks, however, the data will include the values of only some...
Charles Elkan, Russell Greiner
ical reasoning facility that would allow them, for example, to use information about treating one disease to help determine how to treat another. The Cyc project, under the leadership of Douglas...
Adaptive Derivation Processes (1993)
Introduction Many reasoning systems must reach conclusions based on stored information; we can often model this as deriving logical conclusions from a given knowledge base of facts. We of course...
Learning to Classify Incomplete Examples (1993)
Dale Schuurmans, Russell Greiner
Most research on supervised learning assumes the attributes of training and test examples are completely specified. Real-world data, however, is often incomplete. This paper studies the task of...
A formal analysis of solution caching (1992)
Vinay K. Chaudhri, Russell Greiner
Many inference management systems store and maintain the conclusions found during a derivation process in a form that allows these conclusions to be used during subsequent derivations. As this...
Learning Useful Horn Approximations (1992)
Russell Greiner, Dale Schuurmans
While the task of answering queries from an arbitrary propositional theory is intractable in general, it can typically be performed efficiently if the theory is Horn. This suggests that it may be...
A Formal Analysis of Solution Caching (1992)
Vinay K. Chaudhri, Russell Greiner
Many inference management systems store and maintain the conclusions found during a derivation process in a form that allows these conclusions to be used during subsequent derivations. As this...
Probabilistic Hill-Climbing: Theory and Applications (1992)
Many learning systems search through a space of possible performance elements, seeking an element with high expected utility. As the task of finding the globally optimal element is usually...
A Statistical Approach to Solving the EBL Utility Problem (1992)
Russell Greiner, Igor Jurisica
Many "learning from experience" systems use information extracted from problem solving experiences to modify a performance element PE, forming a new element PE 0 that can solve these and...
Learning an Optimally Accurate Representational System (1992)
Russell Greiner, Dale Schuurmans
The multiple extension problem arises because a default theory can use different subsets of its defaults to propose different, mutually incompatible, answers to some queries. This paper presents an...
A Statistical Approach to Solving the EBL Utility Problem (1992)
Russell Greiner, Igor Jurisica
Many "learning from experience" systems use information extracted from problem solving experiences to modify a performance element PE, forming a new element PE 0 that can solve these and...
Learning Efficient Query Processing Strategies (1992)
A query processor qp uses the rules in a rule base to reduce a given query to a series of attempted retrievals from a database of facts. The qp's expected cost is the average time it requires to...
Probably Approximately Optimal Derivation Strategies (1991)
Russell Greiner, Pekka Orponen
An inference graph can have many "derivation strategies", each a particular ordering of the steps involved in reducing a given query to a sequence of database retrievals. An "optimal...
Measuring and Improving the Effectiveness of Representations (1991)
Russell Greiner, Charles Elkan
This report discusses what it means to claim that a representation is an effective encoding of knowledge. We first present dimensions of merit for evaluating representations, based on the view that...
Probabilistic Hill-Climbing (1991)
William W. Cohen, Russell Greiner, Dale Schuurmans
: 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 optimum is...
Measuring and improving the effectiveness of representations (1991)
greinerQcs.toronto.eduAbstract This report discusses what it means to claim that a representation is an effective encoding of knowledge. We first present dimensions of merit for evaluating...
Classical and Logic-Based Dynamic Observers for Finite Automata (1991)
CAINES, PETER E., GREINER, RUSSELL, WANG, SUNING
This paper formulates the state estimation problem for a partially observed input-state-output (N-state) automation in terms of a classical observer automaton each of whose nodes correspond to the...
Probably Approximately Optimal Satisficing Strategies (1990)
Russell Greiner Siemens, Russell Greiner, Pekka Orponen
A satisficing search problem consists of a set of probabilistic experiments to be performed in some order, seeking a satisfying configuration of successes and failures. The expected cost of the...
Probably Approximately Optimal Satisficing Strategies (1990)
Russell Greiner, Pekka Orponen
A satisficing search problem consists of a set of probabilistic experiments to be performed in some order, seeking a satisfying configuration of successes and failures. The expected cost of the...
On The Sample Complexity Of Finding Good Search Strategies (1990)
Pekka Orponen, Russell Greiner
A satisficing search problem consists of a set of probabilistic experiments to be performed in some order, without repetitions, until a satisfying configuration of successes and failures has been...
Finding Optimal Derivation Strategies in Redundant Knowledge Bases (1990)
A backward chaining process uses a collection of rules to reduce a given goal to a sequence of data-base retrievals. A "derivation strategy" is an ordering on these steps, specifying when...
Probably Approximately Optimal Satisficing Strategies (1990)
Russell Greiner, Pekka Orponen
A satisficing search problem consists of a set of probabilistic experiments to be performed in some order, seeking a satisfying configuration of successes and failures. The expected cost of the...
A Correction to the Algorithm in Reiter's Theory of Diagnosis (1989)
Russell Greiner, Barbara A. Smith, Ralph W. Wilkerson
Reiter [1987] has developed a general theory of diagnosis based on first principles. His algorithm computes all diagnoses which explain the differences between the predicted and observed behavior of...
Learning by understanding analogies /--Russell greiner. (1985)
Photocopy of typescript. Ann Arbor, Mich.: University Microfilms, 1987. 22 cm.
Learning by understading analogies /--by Russell Greiner. (1985)
Thesis (Ph. D.)--Stanford University, 1985.
Learning by understanding analogies / (1985)
Thesis (Ph. D.)--Stanford University, 1985.
Learning by understanding analogies / (1985)
Thesis (Ph. D.)--Stanford University, 1985.
Szafron, Duane, Lu, Paul, Greiner, Russell, Wishart, David S., Poulin, Brett, Eisner, Roman, ...
Proteome Analyst (PA) (http://www.cs.ualberta.ca/~bioinfo/PA/) is a publicly available, high-throughput, web-based system for predicting various properties of each protein in an entire proteome....
Lu, Paul, Szafron, Duane, Greiner, Russell, Wishart, David S., Fyshe, Alona, Pearcy, Brandon, ...
PA-GOSUB (Proteome Analyst: Gene Ontology Molecular Function and Subcellular Localization) is a publicly available, web-based, searchable and downloadable database that contains the sequences,...
The Path-A metabolic pathway prediction web server
Pireddu, Luca, Szafron, Duane, Lu, Paul, Greiner, Russell
Pathway Analyst (Path-A) is a publicly available web server () that predicts metabolic pathways. It takes a FASTA format file containing a set of query protein sequences from a single organism (a...
Szafron, Duane, Lu, Paul, Greiner, Russell, Wishart, David S., Poulin, Brett, Eisner, Roman, ...
Proteome Analyst (PA) (http://www.cs.ualberta.ca/~bioinfo/PA/) is a publicly available, high-throughput, web-based system for predicting various properties of each protein in an entire proteome....
Lu, Paul, Szafron, Duane, Greiner, Russell, Wishart, David S., Fyshe, Alona, Pearcy, Brandon, ...
PA-GOSUB (Proteome Analyst: Gene Ontology Molecular Function and Subcellular Localization) is a publicly available, web-based, searchable and downloadable database that contains the sequences,...
The Path-A metabolic pathway prediction web server
Pireddu, Luca, Szafron, Duane, Lu, Paul, Greiner, Russell
Pathway Analyst (Path-A) is a publicly available web server () that predicts metabolic pathways. It takes a FASTA format file containing a set of query protein sequences from a single organism (a...