Papers by Tom Minka (also available by topic)

Expectation propagation
|   Bayesian methods
|   |   Probabilistic modeling
|   |   |   Optimization
|   |   |   |   Text retrieval
|   |   |   |   |   Computer vision
|   |   |   |   |   |

E						2019	From automatic differentiation to message passing
E						2018	TrueSkill 2: An improved Bayesian skill rating system
E						2017	Belief Propagation with Strings
	B					2016	Bayesian prior choice in IRT estimation using MCMC and variational Bayes (code)
E						2015	An introduction to Expectation Propagation (video lecture) (slides)
E						2014	Elastic Distributed Bayesian Collaborative Filtering
	B					2014	A* Sampling
	B					2014	Knowing what we don't know in NCAA Football ratings: Understanding and using structured uncertainty
		M				2013	Detecting Parameter Symmetries in Probabilistic Models
	B					2012	A Bayesian Graphical Model for Adaptive Crowdsourcing and Aptitude Testing
	B					2012	Spot Localization using PHY Layer Information
	B					2011	Non-conjugate Variational Message Passing for Multinomial and Binary Regression
E						2010	Sparse-posterior Gaussian Processes for general likelihoods
	B	M		T		2010	A Novel Click Model and Its Applications to Online Advertising
		M				2009	Probabilistic Programming with Infer.NET
E						2009	Video lectures on Approximate Inference
	B					2009	Automating variational inference for statistics and data mining
E						2009	Virtual Vector Machine for Bayesian Online Classification
	B	M		T		2009	Click Chain Model in Web Search
E		M				2008	Gates: A graphical notation for mixture models
				T		2008	Selection bias in the LETOR datasets
	B				V	2008	Bayesian Color Constancy Revisited
				T		2008	SoftRank: Optimising Non-Smooth Ranking Metrics
E	B					2007	TrueSkill Through Time: Revisiting the History of Chess
		M		T		2007	The Smoothed Dirichlet distribution: A new building block for generative topical models
E	B					2006	TrueSkill: A Bayesian Skill Rating System
			O			2006	Local Training and Belief Propagation
E						2006	Window-based expectation propagation for adaptive signal detection in flat-fading channels
	B		O		V	2006	Cosegmentation of Image Pairs by Histogram Matching - Incorporating a Global Constraint into MRFs
		M			V	2006	Principled Hybrids of Generative and Discriminative Models
		M				2005	Discriminative models, not discriminative training
	B				V	2005	Object Categorization by Learned Universal Visual Dictionary	A Bayesian version of agglomerative information bottleneck.
E						2005	Structured Region Graphs: Morphing EP into GBP
E						2005	Divergence measures and message passing
E	B					2005	Bayesian Conditional Random Fields
E						2004	Power EP
E						2004	A roadmap to research on EP
E	B					2004	Predictive Automatic Relevance Determination by Expectation Propagation	Preventing overfitting in ARD.
		M			V	2004	Exemplar-based likelihoods using the PDF projection theorem	How to properly normalize distributions over image features.
		M		T	V	2003	The `summation hack' as an outlier model	An explanation of a common trick used in computer vision and text retrieval.
E						2003	Tree-structured approximations by expectation propagation
E						2003	Expectation Propagation for Infinite Mixtures
	B				V	2003	Bayesian Color Constancy with Non-Gaussian Models
E						2003	Expectation Propagation for Signal Detection in Flat-fading Channels
		M				2003	Building statistical models by visualization
		M				2003	Conjugate Analysis of the Conway-Maxwell-Poisson Distribution
		M				2003	A Useful Distribution for Fitting Discrete Data: Revival of the COM-Poisson
		M				2003	Computing with the COM-Poisson distribution
			O			2002	Estimating a Gamma distribution
	B					2002	Hessian-based Markov Chain Monte-Carlo Algorithms
E	B					2002	Bayesian inference in dynamic models -- an overview
		M				2002	Judging significance from error bars	Something everyone should know how to do, but probably doesn't.
	B					2002	Bayesian Spectrum Estimation of Unevenly Sampled Nonstationary Data
				T		2002	Novelty and Redundancy Detection in Adaptive Filtering
E				T		2002	Expectation-Propagation for the Generative Aspect Model
			O			2003	A comparison of numerical optimizers for logistic regression	Derives and compares eight methods, including iterative scaling.
E						2001	The EP energy function and minimization schemes
E						2001	Expectation Propagation for approximate Bayesian inference	UAI version of my thesis, with some extra results.
E						2001	A family of algorithms for approximate Bayesian inference	(PhD thesis work) A powerful generalization of belief propagation.
	B					2001	Using lower bounds to approximate integrals	A new interpretation and generalization of Variational Bayes.
			O			2000	Beyond Newton's method	Custom approximations for fast optimization.
			O			2000	Estimating a Dirichlet distribution	Optimization using Newton, modified Newton, and lower bounds.
	B					2000	Automatic choice of dimensionality for PCA
	B					2000	Bayesian model selection
	B					2000	Deriving quadrature rules from Gaussian processes
	B					2000	Distance measures as prior probabilities
	B					2000	Bayesian model averaging is not model combination
	B					2000	Empirical Risk Minimization is an incomplete inductive principle
		M				1999	Learning How to Learn is Learning With Point Sets
	B					1999	Linear regression with errors in both variables: A proper Bayesian approach	Total least squares is not optimal.
		M				1999	The Dirichlet-tree distribution	The next time you use a Dirichlet, consider a Dirichlet-tree instead.
					V	2001	Document image decoding using iterated complete path search
		M				1998	From Hidden Markov Models to Linear Dynamical Systems
	B					1998	Bayesian inference, entropy, and the multinomial distribution	How empirical entropy and empirical mutual information can arise in Bayesian inference.
			O			1998	Expectation-Maximization as lower bound maximization
	B					1998	Bayesian linear regression
	B					1998	Bayesian inference of a uniform distribution	Bayesian methods succeed where maximum-likelihood does not.
	B					1998	Inferring a Gaussian distribution	Bayes provides a new approach to this age-old problem.
		M				1998	Independence Diagrams	A summary of Bayesian network notation.
	B					1998	Pathologies of Orthodox Statistics
		M				1998	Nuances of probability theory
	B				V	1998	An Optimized Interaction Strategy for Bayesian Relevance Feedback
		M				1997	Old and New Matrix Algebra Useful for Statistics
					V	1996	Modeling user subjectivity in image libraries
					V	1996	An Image Database Browser that Learns from User Interaction
					V	1997	Interactive Learning using a "Society of Models"
					V	1995	Vision Texture for Annotation