This paper proposes a novel robust sparse representation method, called the two-stage sparse representation (TSR), for robust recognition on a large-scale database. Based on the divide and conquer strategy, TSR divides the procedure of robust recognition into outlier detection stage and recognition stage. In the first stage, a weighted linear regression is used to learn a metric in which noise and outliers in image pixels are detected. In the second stage, based on the learnt metric, the large-scale dataset is firstly filtered into a small set according to the nearest neighbor criterion. Then a sparse representation is computed by the non-negative least squares technique. The sparse solution is unique and can be optimized efficiently. The extensive numerical experiments on several public databases demonstrate that the proposed TSR approach generally obtains better classification accuracy than the state of the art Sparse Representation Classification (SRC). At the same time, by using the TSR, a significant reduction of computational cost is reached by over fifty times in comparison with the SRC, which enables the TSR to be deployed more suitably for large-scale dataset.

Density modeling is notoriously difficult for high dimensional data. One approach to the problem is to search for a lower dimensional manifold which captures the main characteristics of the data. Recently, the Gaussian Process Latent Variable Model (GPLVM) has successfully been used to find low dimensional manifolds in a variety of complex data. The GPLVM consists of a set of points in a low dimensional latent space, and a stochastic map to the observed space. We show how it can be interpreted as a density model in the observed space. However, the GPLVM is not trained as a density model and therefore yields bad density estimates. We propose a new training strategy and obtain improved generalisation performance and better density estimates in comparative evaluations on several benchmark data sets.

The goal of my work is to detect implicit social ties or closely-linked entities within a data set. In data consisting of people (or other entities) and their affiliations or discrete attributes, we identify unusually similar pairs of people, and we pose the question: Can their similarity be explained by chance, or it is due to a direct (“copying”) relationship between the people? The thesis will explore how to assess this question, and in particular how one’s judgments and confidence depend not only on the two people in question but also on properties of the entire data set. I will provide a framework for solving this problem and experiment with it across multiple synthetic and real-world data sets. My approach requires a model of the copying relationship, a model of independent people, and a method for distinguishing between them. I will focus on two aspects of the problem: (1) choosing background models to fit arbitrary, correlated affiliation data, and (2) understanding how the ability to detect copies is affected by factors like data sparsity and the numbers of people and affiliations, independent of the fit of the models.

bnlearn is an R package which includes several algorithms for learning the structure of Bayesian networks with either discrete or continuous variables. Both constraint-based and score-based algorithms are implemented, and can use the functionality provided by the snow package to improve their performance via parallel computing. Several network scores and conditional independence algorithms are available for both the learning algorithms and independent use. Advanced plotting options are provided by the Rgraphviz package.

Learning, which aims at combining probabilistic graphical Markov networks have proved to be a very useful tool to models with first order logics representations. The represent probability distributions over large domains (see work that we present in this paper has been motivated by for instance, Chapter 8 in (Bishop 2006)). A Markov Network Markov Logic Networks (MLN), introduced in (Richardson is an undirected graphical model, where variables are and Domingos 2006). A Markov Logic Network is defined represented by nodes and features on subsets of variables by a set of weighted first-order formulas.

Many social Web sites allow users to annotate the content with descriptive metadata, such as tags, and more recently also to organize content hierarchically. These types of structured metadata provide valuable evidence for learning how a community organizes knowledge. For instance, we can aggregate many personal hierarchies into a common taxonomy, also known as a folksonomy, that will aid users in visualizing and browsing social content, and also to help them in organizing their own content. However, learning from social metadata presents several challenges: sparseness, ambiguity, noise, and inconsistency. We describe an approach to folksonomy learning based on relational clustering that addresses these challenges by exploiting structured metadata contained in personal hierarchies. Our approach clusters similar hierarchies using their structure and tag statistics, then incrementally weaves them into a deeper, bushier tree. We study folksonomy learning using social metadata extracted from the photo-sharing site Flickr. We evaluate the learned folksonomy quantitatively by automatically comparing it to a reference taxonomy. Our empirical results suggest that the proposed framework, which addresses the challenges listed above, improves on existing folksonomy learning methods.

Traditionally, in the fields of artificial intelligence and robotics, representations of the self have been conspicuously absent. Capabilities of systems are listed explicitly by developers during construction and choices between behavioral options are decided based on search, inference, and planning. In robotics, while knowledge of the external world has often been acquired through experience, knowledge about the robot itself has generally been built in by the designer. Built-in models of the robot's kinematics, physical and sensory capabilities, and other equipment have stood in the place of self-knowledge, but none of these representations offer the flexibility, robustness, and functionality that are present in people. In this work, we seek to emulate forms of self-awareness developed during human infancy in our humanoid robot, Nico. In particular, we are interested in the ability to reason about the robot's embodiment and physical capabilities, with the robot building a model of itself through its experiences.

Reasoning by the usage of mental images has been the subject of much debate in Cognitive Science, especially among the schools of depictive and descriptive imagistic representations. Whether or not reasoning with mental images involves a mechanism or a process different from language based reasoning is an important question. This paper proposes that any theory which aims for a cohesive whole needs to be constrained by neurophysiological data and such data can be obtained by the Clock Drawing Test. The Clock Drawing Test (CDT) is a screening tool for cognitive impairment and can be used as a tool to test resilience of certain factors of visual spatial representations. Thus, it can help to form an empirical case for which factors are prone to debility and which factors are not during the onset and progress of cognitive impairment from a mental representation point of view. This paper presents 50 CDT tests done on patients with cognitive impairment and analyses the results which support the case for a depictive rather than a descriptive theory for imagistic representations. Lastly, this paper proposes that there is some evidence for a more dynamic and distributed nature of representation in the observations which question the above dichotomy and can be partly explained by certain aspects of the connectionist school of thought.

It is still unclear what metacognition is. Two main theories about metacognition are reviewed, each of which claims to provide a better explanation of the phenomenon, while discrediting the other theory as inappropriate. My claim is that in order to do justice to the complex phenomenon of metacognition, we must distinguish two levels of this capacity. It can be shown that each of these theories has been trying to explain only one of the two levels and that, consequently, the conflict between them can be dissolved. Finally, I characterize each level and explain some of their interactions.

Solving real-world problems using symbolic planning often requires a simplified formulation of the original problem, since certain subproblems cannot be represented at all or only in a way leading to inefficiency. For example, manipulation planning may appear as a subproblem in a robotic planning context or a packing problem can be part of a logistics task. In this paper we propose an extension of PDDL for specifying semantic attachments. This allows the evaluation of grounded predicates, the change of fluents and the calculation of durations by externally specified functions. Furthermore, we describe a general schema of integrating semantic attachments into forward-chaining planning systems and report on our experience of adding this extension to the planner Temporal Fast Downward. Finally, we present some preliminary experiments using semantic attachments.