Today's Web-enabled deluge of electronic data calls for automated methods of data analysis. Machine learning provides these, developing methods that can automatically detect patterns in data and then use the uncovered patterns to predict future data. This textbook offers a comprehensive and self-contained introduction to the field of machine learning, based on a unified, probabilistic approach. The coverage combines breadth and depth, offering necessary background material on such topics as probability, optimization, and linear algebra as well as discussion of recent developments in the field, including conditional random fields, L1 regularization, and deep learning. The book is written in an informal, accessible style, complete with pseudo-code for the most important algorithms. All topics are copiously illustrated with color images and worked examples drawn from such application domains as biology, text processing, computer vision, and robotics. Rather than providing a cookbook of different heuristic methods, the book stresses a principled model-based approach, often using the language of graphical models to specify models in a concise and intuitive way. Almost all the models described have been implemented in a MATLAB software package--PMTK (probabilistic modeling toolkit)--that is freely available online. The book is suitable for upper-level undergraduates with an introductory-level college math background and beginning graduate students.

Machine Learning : A Probabilistic Perspective

Monte Carlo methods are revolutionizing the on-line analysis of data in fields as diverse as financial modeling, target tracking and computer vision. These methods, appearing under the names of bootstrap filters, condensation, optimal Monte Carlo filters, particle filters and survival of the fittest, have made it possible to solve numerically many complex, non-standard problems that were previously intractable. This book presents the first comprehensive treatment of these techniques, including convergence results and applications to tracking, guidance, automated target recognition, aircraft navigation, robot navigation, econometrics, financial modeling, neural networks, optimal control, optimal filtering, communications, reinforcement learning, signal enhancement, model averaging and selection, computer vision, semiconductor design, population biology, dynamic Bayesian networks, and time series analysis. This will be of great value to students, researchers and practitioners, who have some basic knowledge of probability. Arnaud Doucet received the Ph. D. degree from the University of Paris-XI Orsay in 1997. From 1998 to 2000, he conducted research at the Signal Processing Group of Cambridge University, UK. He is currently an assistant professor at the Department of Electrical Engineering of Melbourne University, Australia. His research interests include Bayesian statistics, dynamic models and Monte Carlo methods. Nando de Freitas obtained a Ph.D. degree in information engineering from Cambridge University in 1999. He is presently a research associate with the artificial intelligence group of the University of California at Berkeley. His main research interests are in Bayesian statistics and the application of on-line and batch Monte Carlo methods to machine learning. Neil Gordon obtained a Ph.D. in Statistics from Imperial College, University of London in 1993. He is with the Pattern and Information Processing group at the Defence Evaluation and Research Agency in the United Kingdom. His research interests are in time series, statistical data analysis, and pattern recognition with a particular emphasis on target tracking and missile guidance.

/pdf/sequential-monte-carlo-methods-in-practice-3kfxeebn8e.pdf

Sequential Monte Carlo methods in practice

In this book, Andrew Harvey sets out to provide a unified and comprehensive theory of structural time series models. Unlike the traditional ARIMA models, structural time series models consist explicitly of unobserved components, such as trends and seasonals, which have a direct interpretation. As a result the model selection methodology associated with structural models is much closer to econometric methodology. The link with econometrics is made even closer by the natural way in which the models can be extended to include explanatory variables and to cope with multivariate time series. From the technical point of view, state space models and the Kalman filter play a key role in the statistical treatment of structural time series models. The book includes a detailed treatment of the Kalman filter. This technique was originally developed in control engineering, but is becoming increasingly important in fields such as economics and operations research. This book is concerned primarily with modelling economic and social time series, and with addressing the special problems which the treatment of such series poses. The properties of the models and the methodological techniques used to select them are illustrated with various applications. These range from the modellling of trends and cycles in US macroeconomic time series to to an evaluation of the effects of seat belt legislation in the UK.

Forecasting, Structural Time Series Models and the Kalman Filter

This paper presents a tutorial introduction to the use of variational methods for inference and learning in graphical models (Bayesian networks and Markov random fields). We present a number of examples of graphical models, including the QMR-DT database, the sigmoid belief network, the Boltzmann machine, and several variants of hidden Markov models, in which it is infeasible to run exact inference algorithms. We then introduce variational methods, which exploit laws of large numbers to transform the original graphical model into a simplified graphical model in which inference is efficient. Inference in the simpified model provides bounds on probabilities of interest in the original model. We describe a general framework for generating variational transformations based on convex duality. Finally we return to the examples and demonstrate how variational algorithms can be formulated in each case.

/pdf/an-introduction-to-variational-methods-for-graphical-models-4o3vosl5nw.pdf

An introduction to variational methods for graphical models

Dynamic Bayesian Networks: Representation, Inference and Learning by Kevin Patrick Murphy Doctor of Philosophy in Computer Science University of California, Berkeley Professor Stuart Russell, Chair Modelling sequential data is important in many areas of science and engineering. Hidden Markov models (HMMs) and Kalman filter models (KFMs) are popular for this because they are simple and flexible. For example, HMMs have been used for speech recognition and bio-sequence analysis, and KFMs have been used for problems ranging from tracking planes and missiles to predicting the economy. However, HMMs and KFMs are limited in their “expressive power”. Dynamic Bayesian Networks (DBNs) generalize HMMs by allowing the state space to be represented in factored form, instead of as a single discrete random variable. DBNs generalize KFMs by allowing arbitrary probability distributions, not just (unimodal) linear-Gaussian. In this thesis, I will discuss how to represent many different kinds of models as DBNs, how to perform exact and approximate inference in DBNs, and how to learn DBN models from sequential data. In particular, the main novel technical contributions of this thesis are as follows: a way of representing Hierarchical HMMs as DBNs, which enables inference to be done in O(T ) time instead of O(T ), where T is the length of the sequence; an exact smoothing algorithm that takes O(log T ) space instead of O(T ); a simple way of using the junction tree algorithm for online inference in DBNs; new complexity bounds on exact online inference in DBNs; a new deterministic approximate inference algorithm called factored frontier; an analysis of the relationship between the BK algorithm and loopy belief propagation; a way of applying Rao-Blackwellised particle filtering to DBNs in general, and the SLAM (simultaneous localization and mapping) problem in particular; a way of extending the structural EM algorithm to DBNs; and a variety of different applications of DBNs. However, perhaps the main value of the thesis is its catholic presentation of the field of sequential data modelling.

/pdf/dynamic-bayesian-networks-representation-inference-and-yy4fhfxdcj.pdf

Dynamic bayesian networks: representation, inference and learning

Approximating probabilistic inference in Bayesian belief networks is NP-hard

A typical approach to estimate an unknown quantity $\mu$ is to design an experiment that produces a random variable Z, distributed in [0,1] with E[Z]=\mu$, run this experiment independently a number of times, and use the average of the outcomes as the estimate. In this paper, we consider the case when no a priori information about Z is known except that is distributed in [0,1]. We describe an approximation algorithm ${\cal A}{\cal A}$ which, given $\epsilon$ and $\delta$, when running independent experiments with respect to any Z, produces an estimate that is within a factor $1+\epsilon$ of $\mu$ with probability at least $1-\delta$. We prove that the expected number of experiments run by ${\cal A}{\cal A}$ (which depends on Z) is optimal to within a constant factor {for every} Z.

An Optimal Algorithm for Monte Carlo Estimation

Background —Current surgical methods for treating aortic valve and aortic root pathology vary widely, and the basis for selecting one repair or replacement alternative over another continues to evolve. More precise knowledge of the interaction between normal aortic root dynamics and aortic valve mechanics may clarify the implications of various surgical procedures on long-term valve function and durability. Methods and Results —To investigate the role of aortic root dynamics on valve function, we studied the deformation modes of the left, right, and noncoronary aortic root regions during isovolumic contraction, ejection, isovolumic relaxation, and diastole. Radiopaque markers were implanted at the top of the 3 commissures (sinotubular ridge) and at the annular base of the 3 sinuses in 6 adult sheep. After a 1-week recovery, ECG and left ventricular and aortic pressures were recorded in conscious, sedated animals, and the 3D marker coordinates were computed from biplane videofluorograms (60 Hz). Left ventricular preload, contractility, and afterload were independently manipulated to assess the effects of changing hemodynamics on aortic root 3D dynamic deformation. The ovine aortic root undergoes complex, asymmetric deformations during the various phases of the cardiac cycle, including aortoventricular and sinotubular junction strain and aortic root elongation, compression, shear, and torsional deformation. These deformations were not homogeneous among the left, right, and noncoronary regions. Furthermore, changes in left ventricular volume, pressure, and contractility affected the degree of deformation in a nonuniform manner in the 3 regions studied, and these effects varied during isovolumic contraction, ejection, isovolumic relaxation, and diastole. Conclusions —These complex 3D aortic root deformations probably minimize aortic cusp stresses by creating optimal cusp loading conditions and minimizing transvalvular turbulence. Aortic valve repair techniques or methods of replacement using unstented autograft, allograft, or xenograft tissue valves that best preserve this normal pattern of aortic root dynamics should translate into a lower risk of long-term cusp deterioration.

Deformational Dynamics of the Aortic Root Modes and Physiologic Determinants

We have developed a probabilistic forecasting methodology through a synthesis of belief network models and classical time-series analysis. We present the dynamic network model (DNM) and describe methods for constructing, refining, and performing inference with this representation of temporal probabilistic knowledge. The DNM representation extends static belief-network models to more general dynamic forecasting models by integrating and iteratively refining contemporaneous and time-lagged dependencies. We discuss key concepts in terms of a model for forecasting U.S. car sales in Japan.

Dynamic Network Models for Forecasting

We have developed a probabilistic forecasting methodology through a synthesis of belief-network models and classical time-series analysis. We present the dynamic network model (DNM) and describe methods for constructing, refining, and performing inference with this representation of temporal probabilistic knowledge. The DNM representation extends static belief-network models to more general dynamic forecasting models by integrating and iteratively refining contemporaneous and time-lagged dependencies. We discuss key concepts in terms of a model for forecasting U.S. car sales in Japan.

/pdf/dynamic-network-models-for-forecasting-fin97fr6tx.pdf

Paul Dagum

Papers

Approximating probabilistic inference in Bayesian belief networks is NP-hard

An Optimal Algorithm for Monte Carlo Estimation

Deformational Dynamics of the Aortic Root Modes and Physiologic Determinants

Dynamic Network Models for Forecasting

Dynamic network models for forecasting