All intelligence relies on search --- for example, the search for an intelligent agent's next action. Search is only likely to succeed in resource-bounded agents if they have already been biased towards finding the right answer. In artificial agents, the primary source of bias is engineering. This dissertation

A theoretical framework for grounding language is introduced that provides a computational path from sensing and motor action to words and speech acts. The approach combines concepts from semiotics and schema theory to develop a holistic approach to linguistic meaning. Schemas serve as structured beliefs that are grounded in an agent’s physical environment through a causal-predictive cycle of action and perception. Words and basic speech acts are interpreted in terms of grounded schemas. The framework reflects lessons learned from implementations of several language processing robots. It provides a basis for the analysis and design of situated, multimodal communication systems that straddle symbolic and non-symbolic realms.

To operate e#ectively in complex environments learning agents have to selectively ignore irrelevant details by forming useful abstractions. In this article we outline a formulation of abstraction for reinforcement learning approaches to stochastic sequential decision problems modeled as semiMarkov Decision Processes (SMDPs). Building on existing algebraic approaches, we propose the concept of SMDP homomorphism and argue that it provides a useful tool for a rigorous study of abstraction for SMDPs. We apply this framework to di#erent classes of abstractions that arise in hierarchical systems and discuss relativized options, a framework for compactly specifying a related family of temporally-extended actions. Additional details of this work are described in refs. [1, 2, 3].

Intelligent learning environments that support constructivism should provide active learning experiences that are customized for individual learners. To do so, they must determine learner intent and detect misconceptions, and this diagnosis must be performed as non-invasively as possible. To this end, we propose the pedagogical design studio, a design-centered framework for learning environment interfaces. Pedagogical design studios provide learners with a rich, direct manipulation design experience. By exploiting an artifact-based task model that preserves a tight mapping between the interface state and design sub-tasks, they non-invasively infer learners’ intent and detect misconceptions. The task model is then used to tailor problem presentation, produce a customized musical score, and modulate problem-solving intervention. To explore these notions, we have implemented a pedagogical design studio for a constructivist learning environment that provides instruction to middle school students about botanical anatomy and physiology. Evaluations suggest that the design studio framework constitutes an effective approach to interfaces that support constructivist learning.

Design-centered learning environments offer great promise for providing effective, grounded learning experiences. Learners are given a set of design criteria and a library of components which they use to design artifacts that will satisfy the specified criteria. Despite their appeal, design-centered learning environments are plagued with complexities that can overwhelm learners. To address this problem, we have developed a proactive problem-solving focus mechanism that helps learners cope with the complexities inherent in design-centered learning. By exploiting a rich model of the design context, the focus mechanism selects design problems and intervenes with multimedia advice. The mechanism has been implemented in Design-A-Plant, a design-centered learning environment for botanical anatomy and physiology. Formative evaluations with middle school students are encouraging.

One of the defining features of intelligent behaviour is the ordering of individual expressed actions into coherent, apparently rational patterns. Psychology has long assumed that hierarchical and sequential structures internal to the intelligent agent underlie this expression. Recently these assumptions have been challenged by claims that behaviour controlled by such structures is necessarily rigid, brittle, and incapable of reacting quickly and opportunistically to changes in the environment (Hendriks-Jansen 1996, Goldfield 1995, Brooks 1991a). This dissertation is intended to support the hypothesis that sequential and hierarchical structures are necessary to intelligent behaviour, and to refute the above claims of their impracticality. Three forms of supporting evidence are provided: • a demonstration in the form of experimental results in two domains that structured intelligence can lead to robust and reactive behaviour, • a review of recent research results and paradigmatic trends within artificial intelligence, and • a similar examination of related research in natural intelligence.

Abstract—Recently, it has been shown that top-down connections improve recognition in supervised learning. In the work presented here, we show how top-down connections represent temporal context as expectation and how such expectation assists perception in a continuously changing physical world, with which an agent interacts during its developmental learning. In experiments in object recognition and vehicle recognition using two types of networks (which derive either global or local features), it is shown how expectation greatly improves performance, to nearly 100 % after the transition periods. We also analyze why expectation will improve performance in such real world contexts. I.

I argue that data becomes temporarily interesting by itself to some self-improving, but computationally limited, subjective observer once he learns to predict or compress the data in a better way, thus making it subjectively simpler and more beautiful. Curiosity is the desire to create or discover more non-random, nonarbitrary, regular data that is novel and surprising not in the traditional sense of Boltzmann and Shannon but in the sense that it allows for compression progress because its regularity was not yet known. This drive maximizes interestingness, the first derivative of subjective beauty or compressibility, that is, the steepness of the learning curve. It motivates exploring infants, pure mathematicians, composers,

This position paper proposes that the study of embodied cognitive agents, such as humanoid robots, can advance our understanding of the cognitive development of complex sensorimotor, linguistic and social learning skills. This in turn will benefit the design of cognitive robots capable of learning to handle and manipulate objects and tools autonomously, to cooperate and communicate with other robots and humans, and to adapt their abilities to changing internal, environmental, and social conditions. Four key areas of research challenges are discussed, specifically for the issues related to the understanding of: (i) how agents learn and represent compositional actions; (ii) how agents learn and represent compositional lexicons; (iii) the dynamics of social interaction and learning; and (iv) how compositional action and language representations are integrated to bootstrap the cognitive system. The review of specific issues and progress in these areas is then translated into a practical roadmap based on a series of milestones. These milestones provide a possible set of cognitive robotics goals and test-scenarios, thus acting as a research roadmap for future work on cognitive developmental robotics.

Visual occlusion events constitute a major source of depth information. This paper presents a self-organizing neural network that learns to detect, represent, and predict the visibility and invisibility relationships that arise during occlusion events, after a period of exposure to motion sequences containing occlusion and disocclusion events. The network develops two parallel opponent channels or "chains" of lateral excitatory connections for every resolvable motion trajectory. One channel, the "On" chain or "visible" chain, is activated when a moving stimulus is visible. The other channel, the "Off" chain or "invisible" chain, carries a persistent, amodal representation that predicts the motion of a formerly visible stimulus that becomes invisible due to occlusion. The learning rule uses disinhibition from the On chain to trigger learning in the Off chain. The On and Off chain neurons can learn separate associations with object depth ordering. The results are closely related to the ...