https://www2.econ.iastate.edu/tesfatsi/DeepLearningInNeuralNetworksOverview.JSchmidhuber2015.pdf

Deep learning in neural networks

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Data Mining Practical Machine Learning Tools and Techniques

How to Do Things With Words

List of Figures. List of Tables. Preface. Contributing Authors. Series Foreword. Part I: Foundations. 1. An Introduction to Evolutionary Algorithms J.A. Lozano. 2. An Introduction to Probabilistic Graphical Models P. Larranaga. 3. A Review on Estimation of Distribution Algorithms P. Larranaga. 4. Benefits of Data Clustering in Multimodal Function Optimization via EDAs J.M. Pena, et al. 5. Parallel Estimation of Distribution Algorithms J.A. Lozano, et al. 6. Mathematical Modeling of Discrete Estimation of Distribution Algorithms C. Gonzalez, et al. Part II: Optimization. 7. An Empiricial Comparison of Discrete Estimation of Distribution Algorithms R. Blanco., J.A. Lozano. 8. Results in Function Optimization with EDAs in Continuous Domain E. Bengoetxea, et al. 9. Solving the 0-1 Knapsack Problem with EDAs R. Sagarna, P. Larranaga. 10. Solving the Traveling Salesman Problem with EDAs V. Robles, et al. 11. EDAs Applied to the Job Shop Scheduling Problem J.A. Lozano, A. Mendiburu. 12. Solving Graph Matching with EDAs Using a Permutation-Based Representation E. Bengoetxea, et al. Part III: Machine Learning. 13. Feature Subset Selection by Estimation of Distribution Algorithms I. Inza, et al. 14. Feature Weighting for Nearest Neighbor by EDAs I. Inza, et al. 15. Rule Induction by Estimation of Distribution Algorithms B. Sierra, et al. 16. Partial Abductive Inference in Bayesian Networks: An Empirical Comparison Between GAs and EDAs L.M. de Campos, et al.17. Comparing K-Means, GAs and EDAs in Partitional Clustering J. Roure, et al. 18. Adjusting Weights in Artificial Neural Networks using Evolutionary Algorithms C. Cotta, et al. Index.

Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation

In the machine learning field, the performance of a classifier is usually measured in terms of prediction error. In most real-world problems, the error cannot be exactly calculated and it must be estimated. Therefore, it is important to choose an appropriate estimator of the error. This paper analyzes the statistical properties, bias and variance, of the k-fold cross-validation classification error estimator (k-cv). Our main contribution is a novel theoretical decomposition of the variance of the k-cv considering its sources of variance: sensitivity to changes in the training set and sensitivity to changes in the folds. The paper also compares the bias and variance of the estimator for different values of k. The experimental study has been performed in artificial domains because they allow the exact computation of the implied quantities and we can rigorously specify the conditions of experimentation. The experimentation has been performed for two classifiers (naive Bayes and nearest neighbor), different numbers of folds, sample sizes, and training sets coming from assorted probability distributions. We conclude by including some practical recommendation on the use of k-fold cross validation.

Sensitivity Analysis of k-Fold Cross Validation in Prediction Error Estimation

Partial abductive inference in Bayesian networks is intended as the process of generating the J( most probable configurations for a distinguished subset of the network variables (explanation set), given some observations (evidence). This problem, also known as the Maximum a Posteriori Problem, is known to be NP-hard, so exact computation is not always possible. As partial abductive inference in Bayesian networks can be viewed as a combinatorial optimization problem, Genetic Algorithms have been successfully applied to give an approximate algorithm for it (de Campos et al., 1999). In this work we approach the problem by means of Estimation of Distribution Algorithms, and an empirical comparison between the results obtained by Genetic Algorithms and Estimation of Distribution Algorithms is carried out.

Estimation of Distribution Algorithms

http://www.beck-shop.de/fachbuch/inhaltsverzeichnis/9783540230922_TOC_001.pdf

Jose A. Lozano

Papers

Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation

Sensitivity Analysis of k-Fold Cross Validation in Prediction Error Estimation

Estimation of Distribution Algorithms

Parallel Problem Solving from Nature - PPSN VIII

An empirical comparison of four initialization methods for the K-Means algorithm