Fast Learning by Bounding Likelihoods in Sigmoid Type Belief Networks

Sigmoid type belief networks, a class of probabilistic neural networks, provide a natural framework for compactly representing probabilistic information in a variety of unsupervised and supervised learning problems. Often the parameters used in these networks need to be learned from examples. Unfortunately, estimating the parameters via exact probabilistic calculations (i.e, the EM-algorithm) is intractable even for networks with fairly small numbers of hidden units. We propose to avoid the infeasibility of the E step by bounding likelihoods instead of computing them exactly. We introduce extended and complementary representations for these networks and show that the estimation of the network parameters can be made fast (reduced to quadratic optimization) by performing the estimation in either of the alternative domains. The complementary networks can be used for continuous density estimation as well.

Active Learning with Statistical Models

For many types of learners one can compute the statistically 'optimal' way to select data. We review how these techniques have been used with feedforward neural networks. We then show how the same principles may be used to select data for two alternative, statistically-based learning architectures: mixtures of Gaussians and locally weighted regression. While the techniques for neural networks are expensive and approximate, the techniques for mixtures of Gaussians and locally weighted regression are both efficient and accurate.

On Convergence Properties of the EM Algorithm for Gaussian Mixtures

"Expectation-Maximization'' (EM) algorithm and gradient-based approaches for maximum likelihood learning of finite Gaussian mixtures. We show that the EM step in parameter space is obtained from the gradient via a projection matrix $P$, and we provide an explicit expression for the matrix. We then analyze the convergence of EM in terms of special properties of $P$ and provide new results analyzing the effect that $P$ has on the likelihood surface. Based on these mathematical results, we present a comparative discussion of the advantages and disadvantages of EM and other algorithms for the learning of Gaussian mixture models.

Learning from Incomplete Data

Real-world learning tasks often involve high-dimensional data sets with complex patterns of missing features. In this paper we review the problem of learning from incomplete data from two statistical perspectives---the likelihood-based and the Bayesian. The goal is two-fold: to place current neural network approaches to missing data within a statistical framework, and to describe a set of algorithms, derived from the likelihood-based framework, that handle clustering, classification, and function approximation from incomplete data in a principled and efficient manner. These algorithms are based on mixture modeling and make two distinct appeals to the Expectation-Maximization (EM) principle (Dempster, Laird, and Rubin 1977)---both for the estimation of mixture components and for coping with the missing data.

On the Convergence of Stochastic Iterative Dynamic Programming Algorithms

Recent developments in the area of reinforcement learning have yielded a number of new algorithms for the prediction and control of Markovian environments. These algorithms, including the TD(lambda) algorithm of Sutton (1988) and the Q-learning algorithm of Watkins (1989), can be motivated heuristically as approximations to dynamic programming (DP). In this paper we provide a rigorous proof of convergence of these DP-based learning algorithms by relating them to the powerful techniques of stochastic approximation theory via a new convergence theorem. The theorem establishes a general class of convergent algorithms to which both TD(lambda) and Q-learning belong.

Hierarchical Mixtures of Experts and the EM Algorithm

We present a tree-structured architecture for supervised learning. The statistical model underlying the architecture is a hierarchical mixture model in which both the mixture coefficients and the mixture components are generalized linear models (GLIM's). Learning is treated as a maximum likelihood problem; in particular, we present an Expectation-Maximization (EM) algorithm for adjusting the parameters of the architecture. We also develop an on-line learning algorithm in which the parameters are updated incrementally. Comparative simulation results are presented in the robot dynamics domain.

Estimating Dependency Structure as a Hidden Variable

This paper introduces a probability model, the mixture of trees that can account for sparse, dynamically changing dependence relationships. We present a family of efficient algorithms that use EMand the Minimum Spanning Tree algorithm to find the ML and MAP mixtureof trees for a variety of priors, including the Dirichlet and the MDL priors.

Estimating Dependency Structure as a Hidden Variable

This paper introduces a probability model, the mixture of trees that can account for sparse, dynamically changing dependence relationships. We present a family of efficient algorithms that use EM and the Minimum Spanning Tree algorithm to find the ML and MAP mixture of trees for a variety of priors, including the Dirichlet and the MDL priors. We also show that the single tree classifier acts like an implicit feature selector, thus making the classification performance insensitive to irrelevant attributes. Experimental results demonstrate the excellent performance of the new model both in density estimation and in classification.

Àlgebra moderna: conjunts, relacions i aplicacions

Oferim als estudiants universitaris i als lectors interessats aquesta guia didàctica de la matemàtica universitària com a fruit dels nostres anys de docència de les matemàtiques a la Universitat. El resultat final ha esdevingut una col·lecció de setze petits volums agrupats en els dos mòduls d'Àlgebra Lineal i de Càlcul Infinitesimal. El primer volum de la col•lecció, s’inicia amb les nocions primàries del conjunt, element i pertinença que constitueixen el pilar bàsic del llenguatge matemàtic. Tot seguit tractem el tema de les relacions binàries entre els elements d’un conjunt, destacant-hi entre elles les relacions d’equivalència que, com veurem en el proper volum, permetran la fonamentació de les diferents classes de nombres. Finalment, es tracten les aplicacions entre conjunts, un concepte que es desenvoluparà plenament en l’estudi del Càlcul Funcional

Àlgebra moderna: estructures algebraiques

Oferim als estudiants universitaris i als lectors interessats aquesta guia didàctica de la matemàtica universitària com a fruit dels nostres anys de docència de les matemàtiques a la Universitat. El resultat final ha esdevingut una col·lecció de setze petits volums agrupats en els dos mòduls d'Àlgebra Lineal i de Càlcul Infinitesimal. El present volum continua l’estudi de l’Àlgebra moderna iniciada en l’anterior volum. Es comença amb la noció de llei de composició, una operació entre els elements d’un conjunt que utilitzarem pel posterior estudi del concepte d’estructura algebraica, de gran importància en l’Àlgebra moderna. Tot seguit es fa una senzilla introducció a les estructures algebraiques més importants, com són les de grup, anell i cos, fent a més un repàs a les diferents classes de nombres: enters, racionals, reals i complexos

Àlgebra vectorial: aplicacions lineals

Oferim als estudiants universitaris i als lectors interessats aquesta guia didàctica de la matemàtica universitària com a fruit dels nostres anys de docència de les matemàtiques a la Universitat. El resultat final ha esdevingut una col·lecció de setze petits volums agrupats en els dos mòduls d'Àlgebra Lineal i de Càlcul Infinitesimal. En aquest volum es generalitza en primer lloc el concepte d'aplicació entre dos espais vectorials i s'introdueix la important definició d'aplicació lineal. Pel seu estudi s'utilitza l'àlgebra matricial. A continuació es desenvolupen els temes de valors i vectors propis, la diagonalització d'endomorfismes i l'estudi de les formes quadràtiques

Àlgebra vectorial: vectors

Oferim als estudiants universitaris i als lectors interessats aquesta guia didàctica de la matemàtica universitària com a fruit dels nostres anys de docència de les matemàtiques a la Universitat. El resultat final ha esdevingut una col·lecció de setze petits volums agrupats en els dos mòduls d'Àlgebra Lineal i de Càlcul Infinitesimal. Amb aquest sisè volum de la col•lecció iniciem l’estudi de l’Àlgebra vectorial a partir de conceptes propers a la intuïció com són els vectors del pla i de l’espai per, a continuació, fer una generalització del concepte de vector a altres ens matemàtics com polinomis, successions, magnituds econòmiques, etc. En aquest volum utilitzarem sovint la notació matricial, ja coneguda i emprada en volums anteriors, i que esdevé una eina idònia per facilitar la notació dels conceptes i del càlcul entre vectors. Seguim amb l’estudi axiomàtic de l’estructura d’espai vectorial i les seves propietats, que com veurem en el proper volum ens permetrà, entre altres aplicacions a l’economia, deduir els valors i vectors propis d’un endomorfisme i diagonalitzar formes quadràtiques

Active citizenship

The aim of this activity is to allow students to explore the nature of political action, which can be thought of as a form of active as opposed to passive citizenship. By learning about and reflecting upon past instances of political action, or activism, students will be able to start thinking about what is likely to make a campaign successful. It is intended that these reflections can then be applied to their own actions as active citizens. It is hoped that the historical case studies combined with the information provided on different campaigning tools and methods will help to make students feel empowered and inspired to take action. In setting students the task of planning an action, it is expected that time management and organizational skills will be improved. It is believed that by putting themselves in the shoes of activists and going through the process of planning an action, they will have an engaged learning experience. The reflective element of the activity encourages students to form and defend opinions on the relative strengths and weaknesses of different campaigning methods, and on the acceptable limits to political action. This learning activity has been designed presuming no prior knowledge of activism or its methods, and has been successfully used with first year undergraduate students from a variety of disciplines. However, the activity provides a basis for more in-depth study of several issues, or alternatively study into further examples of campaign organizations. There are 3 different learning activities presented on this web site. For a dynamic and well-illustrated introduction to contemporary activism, see Jordan, T. (2002) Activism!: Direct Action, Hacktivism and the Future of Society, London: Reaktion Books Ltd. This material is also available via JORUM.

Exercise sheet 4 pdf

Exercises and solutions in PDF

Exercise sheet 2

Exercises and solutions in LaTex