The paper presents the investigation and implementation of the relationship between diversity and the performance of multiple classifiers on classification accuracy. The study is critical as to build classifiers that are strong and can generalize better. The parameters of the neural network within the committee were varied to induce diversity; hence structural diversity is the focus for this study. The hidden nodes and the activation function are the parameters that were varied. The diversity measures that were adopted from ecology such as Shannon and Simpson were used to quantify diversity. Genetic algorithm is used to find the optimal ensemble by using the accuracy as the cost function. The results observed shows that there is a relationship between structural diversity and accuracy. It is observed that the classification accuracy of an ensemble increases as the diversity increases. There was an increase of 3%-6% in the classification accuracy.

In this paper entropy based methods are compared and used to measure structural diversity of an ensemble of 21 classifiers. This measure is mostly applied in ecology, whereby species counts are used as a measure of diversity. The measures used were Shannon entropy, Simpsons and the Berger Parker diversity indexes. As the diversity indexes increased so did the accuracy of the ensemble. An ensemble dominated by classifiers with the same structure produced poor accuracy. Uncertainty rule from information theory was also used to further define diversity. Genetic algorithms were used to find the optimal ensemble by using the diversity indices as the cost function. The method of voting was used to aggregate the decisions.

This paper discusses the effects of social learning in training of game playing agents. The training of agents in a social context instead of a self-play environment is investigated. Agents that use the reinforcement learning algorithms are trained in social settings. This mimics the way in which players of board games such as scrabble and chess mentor each other in their clubs. A Round Robin tournament and a modified Swiss tournament setting are used for the training. The agents trained using social settings are compared to self play agents and results indicate that more robust agents emerge from the social training setting. Higher state space games can benefit from such settings as diverse set of agents will have multiple strategies that increase the chances of obtaining more experienced players at the end of training. The Social Learning trained agents exhibit better playing experience than self play agents. The modified Swiss playing style spawns a larger number of better playing agents as the population size increases.

The exploration-exploitation dilemma has been an intriguing and unsolved problem within the framework of reinforcement learning. "Optimism in the face of uncertainty" and model building play central roles in advanced exploration methods. Here, we integrate several concepts and obtain a fast and simple algorithm. We show that the proposed algorithm finds a near-optimal policy in polynomial time, and give experimental evidence that it is robust and efficient compared to its ascendants.

In this study, we reproduce two new hybrid intelligent systems, involve three prominent intelligent computing and approximate reasoning methods: Self Organizing feature Map (SOM), Neruo-Fuzzy Inference System and Rough Set Theory (RST),called SONFIS and SORST. We show how our algorithms can be construed as a linkage of government-society interactions, where government catches various states of behaviors: solid (absolute) or flexible. So, transition of society, by changing of connectivity parameters (noise) from order to disorder is inferred.

In this paper, we study the application of sparse principal component analysis (PCA) to clustering and feature selection problems. Sparse PCA seeks sparse factors, or linear combinations of the data variables, explaining a maximum amount of variance in the data while having only a limited number of nonzero coefficients. PCA is often used as a simple clustering technique and sparse factors allow us here to interpret the clusters in terms of a reduced set of variables. We begin with a brief introduction and motivation on sparse PCA and detail our implementation of the algorithm in d'Aspremont et al. (2005). We then apply these results to some classic clustering and feature selection problems arising in biology.

We present an algorithm for effectively generating binary sequences which would be rated by people as highly likely to have been generated by a random process, such as flipping a fair coin.

We obtain a new upper bound on the capacity of a class of discrete memoryless relay channels. For this class of relay channels, the relay observes an i.i.d. sequence $T$, which is independent of the channel input $X$. The channel is described by a set of probability transition functions $p(y|x,t)$ for all $(x,t,y)\in \mathcal{X}\times \mathcal{T}\times \mathcal{Y}$. Furthermore, a noiseless link of finite capacity $R_{0}$ exists from the relay to the receiver. Although the capacity for these channels is not known in general, the capacity of a subclass of these channels, namely when $T=g(X,Y)$, for some deterministic function $g$, was obtained in [1] and it was shown to be equal to the cut-set bound. Another instance where the capacity was obtained was in [2], where the channel output $Y$ can be written as $Y=X\oplus Z$, where $\oplus$ denotes modulo-$m$ addition, $Z$ is independent of $X$, $|\mathcal{X}|=|\mathcal{Y}|=m$, and $T$ is some stochastic function of $Z$. The compress-and-forward (CAF) achievability scheme [3] was shown to be capacity achieving in both cases. Using our upper bound we recover the capacity results of [1] and [2]. We also obtain the capacity of a class of channels which does not fall into either of the classes studied in [1] and [2]. For this class of channels, CAF scheme is shown to be optimal but capacity is strictly less than the cut-set bound for certain values of $R_{0}$. We also evaluate our outer bound for a particular relay channel with binary multiplicative states and binary additive noise for which the channel is given as $Y=TX+N$. We show that our upper bound is strictly better than the cut-set upper bound for certain values of $R_{0}$ but it lies strictly above the rates yielded by the CAF achievability scheme.

Most works related to unithood were conducted as part of a larger effort for the determination of termhood. Consequently, the number of independent research that study the notion of unithood and produce dedicated techniques for measuring unithood is extremely small. We propose a new approach, independent of any influences of termhood, that provides dedicated measures to gather linguistic evidence from parsed text and statistical evidence from Google search engine for the measurement of unithood. Our evaluations revealed a precision and recall of 98.68% and 91.82% respectively with an accuracy at 95.42% in measuring the unithood of 1005 test cases.

This report concerns the problem of dimensionality reduction through information geometric methods on statistical manifolds. While there has been considerable work recently presented regarding dimensionality reduction for the purposes of learning tasks such as classification, clustering, and visualization, these methods have focused primarily on Riemannian manifolds in Euclidean space. While sufficient for many applications, there are many high-dimensional signals which have no straightforward and meaningful Euclidean representation. In these cases, signals may be more appropriately represented as a realization of some distribution lying on a statistical manifold, or a manifold of probability density functions (PDFs). We present a framework for dimensionality reduction that uses information geometry for both statistical manifold reconstruction as well as dimensionality reduction in the data domain.