Showing papers in "Artificial Intelligence in 1995"

TL;DR: By showing that argumentation can be viewed as a special form of logic programming with negation as failure, this paper introduces a general logic-programming-based method for generating meta-interpreters for argumentation systems, a method very much similar to the compiler-compiler idea in conventional programming.

TL;DR: A robust approach to image matching by exploiting the only available geometric constraint, namely, the epipolar constraint, is proposed and a new strategy for updating matches is developed, which only selects those matches having both high matching support and low matching ambiguity.

TL;DR: This model is a hypothesis for primate visual attention, but it also outperforms existing computational solutions for attention in machine vision and is highly appropriate to solving the problem in a robot vision system.

TL;DR: An algorithm based on dynamic programming, which is called Real-Time DP, is introduced, by which an embedded system can improve its performance with experience and illuminate aspects of other DP-based reinforcement learning methods such as Watkins'' Q-Learning algorithm.

TL;DR: This study focusses on interactive processes in the medium and short term: on how agents set up their workplace for particular tasks, and how they continuously manage that workplace.

TL;DR: The success of the approach of building a system with an explicit and grounded representation of cooperative problem solving is used to outline a proposal for the next generation of multi-agent systems.

TL;DR: A general theoretical framework for the synthesis and analysis of autonomous agents is sketched, in which an agent and its environment are modeled as two coupled dynamical systems whose mutual interaction is in general jointly responsible for the agent's behavior.

TL;DR: This paper proposes a general model of social law in a computational system, and investigates some of its properties, and presents an investigation of the automatic synthesis of useful social laws in the framework of a model which refers explicitly to social laws.

TL;DR: Buridan is an implemented least-commitment planner that solves problems of this form and it is proved that the algorithm is both sound and complete, and the interplay between generating plans and evaluating them is described.

TL;DR: An agent architecture is designed and implemented that supports all of these different kinds of adaptation by exploiting a single underlying theoretical concept: an agent dynamically constructs explicit control plans to guide its choices among situation-triggered behaviors.

TL;DR: An approach for integrating planning and control based on behavior schemas, which link physical movements to abstract action descriptions, is proposed, which provides the key to using standard knowledge-based deliberation techniques to generate complex controllers.

TL;DR: A strategic model of negotiation is proposed that takes the passage of time during the negotiation process itself into account, and a distributed negotiation mechanism is introduced that is simple, efficient, stable, and flexible in various situations.

TL;DR: This paper examines how the complexity of domain-independent planning with STRIPS-style operators depends on the nature of the planning operators, and shows conditions under which planning is decidable and undecidable.

TL;DR: A general active vision architecture based on efficiently computable iconic representations, which employs two primary visual routines, one for identifying the visual image near the fovea (object identification), and another for locating a stored prototype on the retina (object location).

TL;DR: It is demonstrated that combining rules and similarities can result in more robust reasoning models, and many seemingly disparate patterns of commonsense reasoning are actually different manifestations of the same underlying process and can be generated using the integrated architecture, which captures the underlying process to a large extent.

TL;DR: A development of a method for tracking visual contours where a training motion of an object can be observed over some extended time and stored as an image sequence is used to learn parameters in a stochastic differential equation model.

TL;DR: The utility of plan-reuse techniques is limited and that these limits have not been determined yet, as it is revealed that plan reuse can actually be strictly more difficult than plan generation.

TL;DR: A set of methods by which a learning agent, called a "critter," can learn a sequence of increasingly abstract and powerful interfaces to control a robot whose sensorimotor apparatus and environment are initially unknown are presented.

TL;DR: This paper develops refinement search semantics for planning, provides a generalized algorithm for refinement planning, and shows that planners that search in the space of (partial) plans are specific instantiations of this algorithm.

TL;DR: The main goal is to define a semantically well-founded logic for approximate reasoning, which is justifiable from the intuitive point of view, and to provide fast algorithms for dealing with it even when using expressive languages.

TL;DR: It is demonstrated that using expectations of object trajectory, size and speed for the particular scene improves robustness and sensitivity in dynamic tracking and segmentation and behavioral evaluation under attentional control using a combination of a static BBN TASKNET and dynamic network.

TL;DR: The initial tasks in this paper are to delimit the boundaries of artificial intelligence, then, to justify calling it a science: is AI science, or is it engineering, or some combination of these?

TL;DR: This work compares configuration of the board-game method to that of a chronological-backtracking problem-solving method for the same application tasks (for example, towers of Hanoi and the Sisyphus room-assignment problem), and examines how method designers can specialize problem-Solving methods by making ontological commitments to certain classes of tasks.

TL;DR: This paper presents a methodology for the design of situated agents that is based on situated-automata theory, which allows designers to describe the informational content of an agent's computational states in a semantically rigorous way without requiring a commitment to conventional run-time symbolic processing.

TL;DR: A restricted set of contraints is identified which gives rise to a class of tractable problems which generalizes the notion of a Horn formula in propositional logic to larger domain sizes, and it is proved that the class of problems generated by any larger set of constraints is NP-complete.

TL;DR: It is shown that the PSPACE-completeness results of Ladner and Halpern and Moses hold for the modal logics K n, T n, S4 n, n ⩾ 1, and K45 n, KD45 n , S5 n , n⩾ 2, even if there is only one primitive proposition in the language.

TL;DR: This paper represents the attempt to characterize a family of complicated and subtle issues concerned with measuring robot task complexity, and develops a theory whereby one sensor can be ''reduced'' to another by adding, deleting, and reallocating (a)-(e) among collaborating autonomous agents.

TL;DR: A key to the success of ADATE is the exact design of these transformations, and how to systematically search for appropriate transformation sequences.

TL;DR: A conceptual framework for computational theories of agents' involvements in their environments is offered, several other fields of research that hold the potential for dialogue with these new computational projects are surveyed, and the principal contributions of the articles are summarized.

TL;DR: Two general approaches to extending reinforcement learning to hidden state tasks are examined, which unifies recent approaches such as the Lion algorithm, the G-algorithm, and CS-QL and assumes that, by appropriate control of perception, the external states can be identified at each point in time from the immediate sensory inputs.