James B. Marshall

Bringing up robot: Fundamental mechanisms for creating a self-motivated, self-organizing architecture

We propose an intrinsic developmental algorithm that is designed to allow a mobile robot to incrementally progress through levels of increasingly sophisticated behavior. We believe that the core ingredients for such a developmental algorithm are abstractions, anticipations, and self-motivations. We describe a multilevel, cascaded discovery and control architecture that includes these core ingredients. As a first step toward implementing the proposed architecture, we explore two novel mechanisms: a governor for automatically regulating the training of a neural network and a path-planning neural network driven by patterns of “mental states” that represent protogoals.

Interactional Motivation in artificial systems: Between extrinsic and intrinsic motivation

This paper introduces Interactional Motivation (IM) as a way to implement self-motivation in artificial systems. An interactionally motivated agent selects behaviors for the sake of enacting the behavior itself rather than for the value of the behaviors outcome. IM contrasts with extrinsic motivation by the fact that it defines the agents motivation independently from the environments state. Because IM does not refer to the environments states, we argue that IM is a form of self-motivation on the same level as intrinsic motivation. IM, however, differs from intrinsic motivation by the fact that IM allows specifying the agents inborn value system explicitly. This paper introduces a formal definition of the IM paradigm and compares it to the reinforcement-learning paradigm as traditionally implemented in Partially Observable Markov Decision Processes (POMDPs).

A self-watching model of analogy-making and perception

This paper describes Metacat, an extension of the Copycat model of analogy-making. The development of Copycat focused on modelling context-sensitive concepts and the ways in which they interact with perception within an abstract microworld of analogy problems. This approach differs from most other models of analogy in its insistence that concepts acquire their semantics from within the system itself, through perception, rather than being imposed from the outside. The present work extends these ideas by incorporating self-perception, episodic memory, and reminding into the model. These mechanisms enable Metacat to explain the similarities and differences that it perceives between analogies, and to monitor and respond to patterns that occur in its own behaviour as it works on analogy problems. This introspective capacity overcomes several limitations inherent in the earlier model, and affords the program a powerful degree of self-control. Metacats architecture includes aspects of both symbolic and connectionist systems. The paper outlines the principal components of the architecture, analyses several sample runs and examples of program-generated commentary about analogies, and discusses Metacats relation to some other well-known models of analogy.

Demonstrating sensemaking emergence in artificial agents: A method and an example

We propose an experimental method to study the possible emergence of sensemaking in artificial agents. This method involves analyzing the agents behavior in a test bed environment that presents regularities in the possibilities of interaction afforded to the agent, while the agent has no presuppositions about the underlying functioning of the environment that explains such regularities. We propose a particular environment that permits such an experiment, called the Small Loop Problem. We argue that the agents behavior demonstrates sensemaking if the agent learns to exploit regularities of interaction to fulfill its self-motivation as if it understood (at least partially) the underlying functioning of the environment. As a corollary, we argue that sensemaking and self-motivation come together. We propose a new method to generate self-motivation in an artificial agent called interactional motivation. An interactionally motivated agent seeks to perform interactions with predefined positive values and avoid interactions with predefined negative values. We applied the proposed sensemaking emergence demonstration method to an agent implemented previously, and produced example reports that suggest that this agent is capable of a rudimentary form of sensemaking.

Calico: a multi-programming-language, multi-context framework designed for computer science education

The Calico project is a multi-language, multi-context programming framework and learning environment for computing education. This environment is designed to support several interoperable programming languages (including Python, Scheme, and a visual programming language), a variety of pedagogical contexts (including scientific visualization, robotics, and art), and an assortment of physical devices (including different educational robotics platforms and a variety of physical sensors). In addition, the environment is designed to support collaboration and modern, interactive learning. In this paper we describe the Calico project, its design and goals, our prototype system, and its current use.

Computational Notebooks for AI Education

Computational notebooks are documents that serve dual purposes: they serve as an archive format containing code, text, images and equations; but they can also be run like computer programs. This paper explores the use of these new computational notebooks to teach AI and introduces tools that we have developed — ICalico and Calysto — to facilitate that use. Not only do these new tools broaden the languages and contexts available to students exploring notebook-based AI computing, but they offer a new mode of teaching and learning for the AI classroom.

Early-stage vision of composite scenes for spatial learning and navigation

Developmental theories suggest that cognitive agents develop through an initial sensorimotor stage during which they learn sequential and spatial regularities. We implemented these views in a computer simulation. Following its intrinsic motivations, the agent autonomously learns sensorimotor contingencies and discovers permanent landmarks by which to navigate in the environment. Besides illustrating developmental theories, this model suggests new ways to implement vision and navigation in artificial systems. Specifically, we coupled a sequence learning mechanism with a visual system capable of interpreting composite visual scenes by inhibiting items that are irrelevant to the agents current motivational state.

The Small Loop Problem: A Challenge for Artificial Emergent Cognition

We propose the Small Loop Problem as a challenge for biologically inspired cognitive architectures. This challenge consists of designing an agent that would autonomously organize its behavior through interaction with an initially unknown environment that offers basic sequential and spatial regularities. The Small Loop Problem demonstrates four principles that we consider crucial to the implementation of emergent cognition: environment-agnosticism, self-motivation, sequential regularity learning, and spatial regularity learning. While this problem is still unsolved, we report partial solutions that suggest that its resolution is realistic.