We address the issue of knots selection for Gaussian predictive process methodology. Predictive process approximation provides an effective solution to the cubic order computational complexity of Gaussian process models. This approximation crucially depends on a set of points, called knots, at which the original process is retained, while the rest is approximated via a deterministic extrapolation. Knots should be few in number to keep the computational complexity low, but provide a good coverage of the process domain to limit approximation error. We present theoretical calculations to show that coverage must be judged by the canonical metric of the Gaussian process. This necessitates having in place a knots selection algorithm that automatically adapts to the changes in the canonical metric affected by changes in the parameter values controlling the Gaussian process covariance function. We present an algorithm toward this by employing an incomplete Cholesky factorization with pivoting and dynamic stopping. Although these concepts already exist in the literature, our contribution lies in unifying them into a fast algorithm and in using computable error bounds to finesse implementation of the predictive process approximation. The resulting adaptive predictive process offers a substantial automatization of Guassian process model fitting, especially for Bayesian applications where thousands of values of the covariance parameters are to be explored.

We extend the potential-based shaping method from Markov decision processes to multi-player general-sum stochastic games. We prove that the Nash equilibria in a stochastic game remains unchanged after potential-based shaping is applied to the environment. The property of policy invariance provides a possible way of speeding convergence when learning to play a stochastic game.

A nonparametric Bayesian extension of Factor Analysis (FA) is proposed where observed data $\mathbf{Y}$ is modeled as a linear superposition, $\mathbf{G}$, of a potentially infinite number of hidden factors, $\mathbf{X}$. The Indian Buffet Process (IBP) is used as a prior on $\mathbf{G}$ to incorporate sparsity and to allow the number of latent features to be inferred. The model's utility for modeling gene expression data is investigated using randomly generated data sets based on a known sparse connectivity matrix for E. Coli, and on three biological data sets of increasing complexity.

Automating the design of heuristic search methods is an active research field within computer science, artificial intelligence and operational research. In order to make these methods more generally applicable, it is important to eliminate or reduce the role of the human expert in the process of designing an effective methodology to solve a given computational search problem. Researchers developing such methodologies are often constrained on the number of problem domains on which to test their adaptive, self-configuring algorithms; which can be explained by the inherent difficulty of implementing their corresponding domain specific software components. This paper presents HyFlex, a software framework for the development of cross-domain search methodologies. The framework features a common software interface for dealing with different combinatorial optimisation problems, and provides the algorithm components that are problem specific. In this way, the algorithm designer does not require a detailed knowledge the problem domains, and thus can concentrate his/her efforts in designing adaptive general-purpose heuristic search algorithms. Four hard combinatorial problems are fully implemented (maximum satisfiability, one dimensional bin packing, permutation flow shop and personnel scheduling), each containing a varied set of instance data (including real-world industrial applications) and an extensive set of problem specific heuristics and search operators. The framework forms the basis for the first International Cross-domain Heuristic Search Challenge (CHeSC), and it is currently in use by the international research community. In summary, HyFlex represents a valuable new benchmark of heuristic search generality, with which adaptive cross-domain algorithms are being easily developed, and reliably compared.

Learning to operate a vehicle is generally accomplished by forming a new cognitive map between the body motions and extrapersonal space. Here, we consider the challenge of remapping movement-to-space representations in survivors of spinal cord injury, for the control of powered wheelchairs. Our goal is to facilitate this remapping by developing interfaces between residual body motions and navigational commands that exploit the degrees of freedom that disabled individuals are most capable to coordinate. We present a new framework for allowing spinal cord injured persons to control powered wheelchairs through signals derived from their residual mobility. The main novelty of this approach lies in substituting the more common joystick controllers of powered wheelchairs with a sensor shirt. This allows the whole upper body of the user to operate as an adaptive joystick. Considerations about learning and risks have lead us to develop a safe testing environment in 3D Virtual Reality. A Personal Augmented Reality Immersive System (PARIS) allows us to analyse learning skills and provide users with an adequate training to control a simulated wheelchair through the signals generated by body motions in a safe environment. We provide a description of the basic theory, of the development phases and of the operation of the complete system. We also present preliminary results illustrating the processing of the data and supporting of the feasibility of this approach.

High dimensional time series are endemic in applications of machine learning such as robotics (sensor data), computational biology (gene expression data), vision (video sequences) and graphics (motion capture data). Practical nonlinear probabilistic approaches to this data are required. In this paper we introduce the variational Gaussian process dynamical system. Our work builds on recent variational approximations for Gaussian process latent variable models to allow for nonlinear dimensionality reduction simultaneously with learning a dynamical prior in the latent space. The approach also allows for the appropriate dimensionality of the latent space to be automatically determined. We demonstrate the model on a human motion capture data set and a series of high resolution video sequences.

Furthermore, interaction with such systems has to happen in a very specific and narrow spectrum of interfaces, with limited margin of flexibility and adaptability. BRIEF SUMMARY Harmonic theory provides a mathematical framework to describe the structure, behavior, evolution and emergence of harmonic systems. A harmonic system is context aware, contains elements that manifest characteristics either collaboratively or independently according to system's expression and can interact with its environment. This theory provides a fresh way to analyze emergence and collaboration of "ad-hoc" and complex systems.

The Beth definability property, a well-known property from classical logic, is investigated in the context of description logics (DLs): if a general L-TBox implicitly defines an L-concept in terms of a given signature, where L is a DL, then does there always exist over this signature an explicit definition in L for the concept? This property has been studied before and used to optimize reasoning in DLs. In this paper a complete classification of Beth definability is provided for extensions of the basic DL ALC with transitive roles, inverse roles, role hierarchies, and/or functionality restrictions, both on arbitrary and on finite structures. Moreover, we present a tableau-based algorithm which computes explicit definitions of at most double exponential size. This algorithm is optimal because it is also shown that the smallest explicit definition of an implicitly defined concept may be double exponentially long in the size of the input TBox. Finally, if explicit definitions are allowed to be expressed in first-order logic then we show how to compute them in EXPTIME.

This paper addresses one of the key components of the mining process: the geological prediction of natural resources from spatially distributed measurements. We present a novel approach combining undirected graphical models with ensemble classifiers to provide 3D geological models from multiple sensors installed in an autonomous drill rig. Drill sensor measurements used for drilling automation, known as measurement-while-drilling (MWD) data, have the potential to provide an estimate of the geological properties of the rocks being drilled. The proposed method maps MWD parameters to rock types while considering spatial relationships, i.e., associating measurements obtained from neighboring regions. We use a conditional random field with local information provided by boosted decision trees to jointly reason about the rock categories of neighboring measurements. To validate the approach, MWD data was collected from a drill rig operating at an iron ore mine. Graphical models of the 3D structure present in real data sets possess a high number of nodes, edges and cycles, making them intractable for exact inference. We provide a comparison of three approximate inference methods to calculate the most probable distribution of class labels. The empirical results demonstrate the benefits of spatial modeling through graphical models to improve classification performance.

Socially Assistive Robotics (SAR) defines the research regarding robots which provide assistance to users through social interaction. Socially assistive robots are being studied for therapeutic use with children with autism spectrum disorders (ASD). It has been observed that children with ASD interact with robots differently than with people or toys. This may indicate an intrinsic interest in such machines, which could be applied as a robot augmentation for an intervention for children with ASD. Preliminary studies suggest that robots may act as intrinsically-rewarding social partners for children with autism. However, enabling a robot to understand social behavior, and do so while interacting with the child, is a challenging problem. Children are highly individual and thus technology used for social interaction requires recognition of a wide-range of social behavior. This work addresses the challenge of designing behaviors for socially assistive robots in order to enable them to recognize and appropriately respond to a child’s free-form behavior in unstructured play contexts. The focus on free-form behavior is inspired by and grounded in existing approaches to therapeutic intervention with children with ASD. This model emphasizes creating circles of communication and fostering engagement through play. A key aspect of this approach is to recognize social behavior and use “engagements” to bolster social interaction behavior, and to study the ethical implications of therapeutic robotics applications.