A procedure for unfolding the true distribution from experimental data is presented. Machine learning methods are applied for simultaneous identification of an apparatus function and solving of an inverse problem. A priori information about the true distribution from theory or previous experiments is used for Monte-Carlo simulation of the training sample. The training sample can be used to calculate a transformation from the true distribution to the measured one. This transformation provides a robust solution for an unfolding problem with minimal biases and statistical errors for the set of distributions used to create the training sample. The dimensionality of the solved problem can be arbitrary. A numerical example is presented to illustrate and validate the procedure.

We address the online linear optimization problem when the actions of the forecaster are represented by binary vectors. Our goal is to understand the magnitude of the minimax regret for the worst possible set of actions. We study the problem under three different assumptions for the feedback: full information, and the partial information models of the so-called "semi-bandit", and "bandit" problems. We consider both $L_\infty$-, and $L_2$-type of restrictions for the losses assigned by the adversary. We formulate a general strategy using Bregman projections on top of a potential-based gradient descent, which generalizes the ones studied in the series of papers Gyorgy et al. (2007), Dani et al. (2008), Abernethy et al. (2008), Cesa-Bianchi and Lugosi (2009), Helmbold and Warmuth (2009), Koolen et al. (2010), Uchiya et al. (2010), Kale et al. (2010) and Audibert and Bubeck (2010). We provide simple proofs that recover most of the previous results. We propose new upper bounds for the semi-bandit game. Moreover we derive lower bounds for all three feedback assumptions. With the only exception of the bandit game, the upper and lower bounds are tight, up to a constant factor. Finally, we answer a question asked by Koolen et al. (2010) by showing that the exponentially weighted average forecaster is suboptimal against $L_{\infty}$ adversaries.

We develop a coherent framework for integrative simultaneous analysis of the exploration-exploitation and model order selection trade-offs. We improve over our preceding results on the same subject (Seldin et al., 2011) by combining PAC-Bayesian analysis with Bernstein-type inequality for martingales. Such a combination is also of independent interest for studies of multiple simultaneously evolving martingales.

In this paper, we propose to (seamlessly) integrate b-bit minwise hashing with linear SVM to substantially improve the training (and testing) efficiency using much smaller memory, with essentially no loss of accuracy. Theoretically, we prove that the resemblance matrix, the minwise hashing matrix, and the b-bit minwise hashing matrix are all positive definite matrices (kernels). Interestingly, our proof for the positive definiteness of the b-bit minwise hashing kernel naturally suggests a simple strategy to integrate b-bit hashing with linear SVM. Our technique is particularly useful when the data can not fit in memory, which is an increasingly critical issue in large-scale machine learning. Our preliminary experimental results on a publicly available webspam dataset (350K samples and 16 million dimensions) verified the effectiveness of our algorithm. For example, the training time was reduced to merely a few seconds. In addition, our technique can be easily extended to many other linear and nonlinear machine learning applications such as logistic regression.

Originating from Allen's Interval Algebra, composition-based reasoning has been widely acknowledged as the most popular reasoning technique in qualitative spatial and temporal reasoning. Given a qualitative calculus (i.e. a relation model), the first thing we should do is to establish its composition table (CT). In the past three decades, such work is usually done manually. This is undesirable and error-prone, given that the calculus may contain tens or hundreds of basic relations. Computing the correct CT has been identified by Tony Cohn as a challenge for computer scientists in 1995. This paper addresses this problem and introduces a semi-automatic method to compute the CT by randomly generating triples of elements. For several important qualitative calculi, our method can establish the correct CT in a reasonable short time. This is illustrated by applications to the Interval Algebra, the Region Connection Calculus RCC-8, the INDU calculus, and the Oriented Point Relation Algebras. Our method can also be used to generate CTs for customised qualitative calculi defined on restricted domains.

We present two alternative ways to apply PAC-Bayesian analysis to sequences of dependent random variables. The first is based on a new lemma that enables to bound expectations of convex functions of certain dependent random variables by expectations of the same functions of independent Bernoulli random variables. This lemma provides an alternative tool to Hoeffding-Azuma inequality to bound concentration of martingale values. Our second approach is based on integration of Hoeffding-Azuma inequality with PAC-Bayesian analysis. We also introduce a way to apply PAC-Bayesian analysis in situation of limited feedback. We combine the new tools to derive PAC-Bayesian generalization and regret bounds for the multiarmed bandit problem. Although our regret bound is not yet as tight as state-of-the-art regret bounds based on other well-established techniques, our results significantly expand the range of potential applications of PAC-Bayesian analysis and introduce a new analysis tool to reinforcement learning and many other fields, where martingales and limited feedback are encountered.

We study the convergence of Markov Decision Processes made of a large number of objects to optimization problems on ordinary differential equations (ODE). We show that the optimal reward of such a Markov Decision Process, satisfying a Bellman equation, converges to the solution of a continuous Hamilton-Jacobi-Bellman (HJB) equation based on the mean field approximation of the Markov Decision Process. We give bounds on the difference of the rewards, and a constructive algorithm for deriving an approximating solution to the Markov Decision Process from a solution of the HJB equations. We illustrate the method on three examples pertaining respectively to investment strategies, population dynamics control and scheduling in queues are developed. They are used to illustrate and justify the construction of the controlled ODE and to show the gain obtained by solving a continuous HJB equation rather than a large discrete Bellman equation.

Various strategies and techniques have been proposed for the generation of questions/answers tests in Intelligent Tutoring Systems by using OWL (Web Ontology Language) ontolo- gies. Currently there have been no known methods to utilize SWRL rules for this task. This paper presents a system and a set of strategies that can be used in order to automatically generate multiple choice questions from SWRL rules. The aim of the proposed framework is to support further research in the area and to be a testbed for the development of more advanced assessment techniques.

Multiagent planning is computationally hard in the general case due to the exponential blowup in the action space induced by concurrent action of different agents. At the same time, many scenarios require the computation of plans that are strategically meaningful for self-interested agents, in order to ensure that there would be sufficient incentives for those agents to participate in a joint plan. In this paper, we present a multiagent planning and plan improvement method that is based on conducting iterative best-response planning using standard single-agent planning algorithms. In constrained types of planning scenarios that correspond to congestion games, this is guaranteed to converge to a plan that is a Nash equilibrium with regard to agents' preference profiles over the entire plan space. Our empirical evaluation beyond these restricted scenarios shows, however, that the algorithm has much broader applicability as a method for plan improvement in general multiagent planning problems. Extensive empirical experiments in various domains illustrate the scalability of our method in both cases.

Closely related genres have complex interrelations. An antecedent genre can constrain a subsequent genre, but changing rhetorical situations can lead to distinctions between an antecedent and its descendent. In this study, we assess two genres of detective fiction to determine their hierarchical relation to one another. We use the Gramulator, a computational tool that identifies indicative lexical features, to explain the relationship between whodunit fiction and hardboiled fiction . We conclude, based on the indicative lexical features of the expositions in texts, that the two are sibling genres.