In many spatial problems, it can be difficult to create a state representation that is abstract enough so that irrelevant details are ignored, but also accurate enough so that important states of the problem can be differentiated. This is especially difficult for agents that address a variety of problems. A potential way to resolve this difficulty is by using two representations of the spatial state of the problem: one abstract and one concrete, along with internal (imagery) operations that modify the concrete representation based on the contents of the abstract representation. In this paper, we argue that such a system can allow plans and policies to be expressed that can better solve a wider class of problems than would otherwise be possible. An example of such a plan is described. The theoretical aspects of what imagery is, how it differs from other techniques, and why it provides a benefit are explored.

Multi-representational architectures exploit diversity to yield the breadth of capabilities required for intelligent behavior in the world, but in so doing can sacrifice too much of the complementary benefits of architectural uniformity. The proposal here is to couple the benefits of diversity and uniformity through establishment of a uniform graph-based implementation level for diverse architectures.

In this position paper we discuss some general properties involved in high-level cognition in situated agents, and sketch an architecture relating sub-symbolic and symbolic information, in which the notion of perceptual inference plays a central role.

In earlier work, we have shown how a cognitive architecture can be augmented with a diagrammatic reasoning system to produce a bimodal cognitive architecture. In this paper, we show how this bimodal architecture is also bi-representational (multi-representational in the general case) by describing a desiderata for representational formalisms and showing how the diagrammatic representation in biSoar satisfies these requirements.

The Raven's Progressive Matrices (RPM) test is a commonly used test of general human intelligence. The RPM is somewhat unique as a general intelligence test in that it focuses on visual problem solving, and in particular, on visual similarity and analogy. We are developing a small set of methods for problem solving in the RPM which use propositional, imagistic, and multimodal representations, respectively, to investigate how different representations can contribute to visual problem solving and how the effects of their use might emerge in behavior.

Comirit is a framework for commonsense reasoning that combines simulation, logical deduction and passive machine learning. While a passive, observation-driven approach to learning is safe and highly conservative, it is limited to inte-raction only with those objects that it has previously ob-served. In this paper we describe a preliminary exploration of methods for extending Comirit to allow safe action selection in uncertain situations, and to allow reward-maximizing selection of behaviors.

A sizable fraction of current research into human visuo-spatial knowledge processing explicitly or implicitly suggests a spatial processing of certain knowledge types and a visual processing of others. Similarly, many formal and technical approaches for representing and processing visuo-spatial information in artificial intelligence, in computational cognitive modeling, or in knowledge representation and reasoning explicitly or implicitly treat visual and spatial information as belonging to separate types. While there exists good evidence for some differences in mental processing of different visuo-spatial knowledge types, there is much less reason to maintain the currently ascribed separation between the visual and the spatial. We provide arguments on why strict dichotomies seem unwarranted with regard to descriptions of human mental spatial reasoning and disadvantageous for the formal and technical approaches. We build upon a synopsis of psychological evidence for the existence of multiple knowledge type specific representations in human visuo-spatial reasoning and discuss the notion of scalable representation structures. In absence of proof to the contrary, it seems better practice to assume that (a) many of the type differences attributed to visuo-spatial knowledge processing are gradual rather than qualitative in nature, and that (b) tasks involving visuo-spatial knowledge of several types are often mentally processed through dynamic associations of structures for processing basal knowledge types. The paper calls for more investigations of human reasoning in visuo-spatial tasks in which knowledge types dynamically change during reasoning. It outlines a research framework for systematically investigating different basal visuo-spatial knowledge types and their combinations with regard to cognitive and computational plausibility. Current research is related to the framework, including research on Casimir, our computational cognitive architecture for reasoning with visuo-spatial knowledge. We argue that a more systematic course of research along the lines of the proposed framework will not only lead to more appropriate descriptions of human cognition (regarding visuo-spatial knowledge processing) but may also spawn more integrated and versatile formal and technical approaches for dealing with visuo-spatial information.

Manifold alignment has been found to be useful in many fields of machine learning and data mining. In this paper we summarize our work in this area and introduce a general framework for manifold alignment. This framework generates a family of approaches to align manifolds by simultaneously matching the corresponding instances and preserving the local geometry of each given manifold. Some approaches like semi-supervised alignment and manifold projections can be obtained as special cases. Our framework can also solve multiple manifold alignment problems and be adapted to handle the situation when no correspondence information is available. The approaches are described and evaluated both theoretically and experimentally, providing results showing useful knowledge transfer from one domain to another. Novel applications of our methods including identification of topics shared by multiple document collections, and biological structure alignment are discussed in the paper.

Face recognition under variable lighting conditions is recognized as one of the most problematic are of the recognition domain by various authors. Previous work suggested that image variations caused by parameters such as illumination, can be modeled by low dimensional subspaces. In this work, we propose a new scheme for recognition under a single variation. Using a generic manifold learning technique like LPP, we are able to construct coordinate systems for the underlying subspace with the help of an optimization step. We performed experiments with face recognition under changing illumination conditions.

Modern mobile robots navigate uncertain environments using complex compositions of camera, laser, and sonar sensor data. Manual calibration of these sensors is a tedious process that involves determining sensor behavior, geometry and location through model specification and system identification. Instead, we seek to automate the construction of sensor model geometry by mining uninterpreted sensor streams for regularities. Manifold learning methods are powerful techniques for deriving sensor structure from streams of sensor data. In recent years, the proliferation of manifold learning algorithms has led to a variety of choices for autonomously generating models of sensor geometry. We present a series of comparisons between different manifold learning methods for discovering sensor geometry for the specific case of a mobile robot with a variety of sensors. We also explore the effect of control laws and sensor boundary size on the efficacy of manifold learning approaches. We find that "motor babbling" control laws generate better geometric sensor maps than mid-line or wall following control laws and identify a novel method for distinguishing boundary sensor elements. We also present a new learning method, sensorimotor embedding, that takes advantage of the controllable nature of robots to build sensor maps.