Asimovian adaptive agents

  title={Asimovian adaptive agents},
  author={D. F. Gordon},
  journal={Journal of Artificial Intelligence Research},
  • D. Gordon
  • Published 1 August 2000
  • Computer Science
  • Journal of Artificial Intelligence Research
The goal of this research is to develop agents that are adaptive and predictable and timely. At first blush, these three requirements seem contradictory. For example, adaptation risks introducing undesirable side effects, thereby making agents' behavior less predictable. Furthermore, although formal verification can assist in ensuring behavioral predictability, it is known to be time-consuming. Our solution to the challenge of satisfying all three requirements is the following. Agents have… 

Assuring the Behavior of Adaptive Agents

Formal Approaches to Agent-Based Systems

A range of agent implementation technologies are reviewed according to five user-based criteria and via a comparison with object-oriented programming. The comparison with OO shows that some parts of

APT Agents: Agents That Are Adaptive, Predictable, and Timely

This paper introduces a paradigm in which all three requirements of adaptability to unforeseen conditions, behavioral assurance, and timeliness of agent responses are simultaneously satisfied, and presents incremental algorithms that can substantially improve the efficiency of reverification.

Verification and repair of control policies for safe reinforcement learning

This work proposes a general-purpose automated methodology based on probabilistic model checking algorithms and tools to verify risk bounds and repair policies and shows that this methodology is also more effective than comparable ones.


This work presents a Bayesian approach to learning exible safety constraints and subsequently verifying whether plans satisfy these constraints, and discusses the advantages and empirical results on an Airspace Planning and Deconiction Task.

Formal Specification and Analysis of Autonomous Systems under Partial Compliance

A technique to formally define system attributes that can change or be changed to deal with dynamic and uncertain environments (denominated weakened specifications) as a partially ordered lattice, and to automatically explore the system under different specifications, using probabilistic model checking, to find the likelihood of satisfying a requirement.

Close Engagements with Artificial Companions: Key Social, Psychological, Ethical, and Design Issues Yorick Wilks (editor) (University of Oxford) Amsterdam: John Benjamins Publishing Company (Natural Language Processing series, edited by Ruslan Mitkov, volume 8), 2010, xxii+315 pp; hardbound, ISBN 97

This book is an edited collection of chapters on artificial companions that purports to discuss the philosophical and ethical issues associated with ACs, what ACs should be like and how to construct them, and to provide examples of special-purpose ACs.

Safe Learning with Real-Time Constraints: A Case Study

This work proposes a solution using automatatheoretic formalisms and associated verification tools, showing experimentally that the approach can yield safety without heavily compromising effectiveness.

Setting Decision Process Optimization into Stochastic vs. Petri Nets Contexts

A new modeling paradigm for developing decision process representation for shortest-path problem and games uses a Lyapunov-like function, and it is shown that the Lyap unov equilibrium point coincides with the Nash equilibrium point under certain restrictions.



Boolean algebra

On a Decision Method in Restricted Second Order Arithmetic

Let SC be the interpreted formalism which makes use of individual variables t, x, y, z,... ranging over natural numbers, monadic predicate variables q( ), r( ), s( ), i( ),... ranging over arbitrary

Version spaces: an approach to concept learning.

The version space approach has been implemented as one component of the Meta-DENDRAL program for learning production rules in the domain of chemical spectroscopy and proofs are given for the correctness of the method for representing version spaces, and of the associated concept learning algorithm, for any countably infinite concept description language.

Adaptive supervisory control of interconnected discrete event systems

  • D. GordonK. Kiriakidis
  • Computer Science
    Proceedings of the 2000. IEEE International Conference on Control Applications. Conference Proceedings (Cat. No.00CH37162)
  • 2000
This paper proposes a novel approach to adaptive supervision for adaptive systems, which aims to address the challenges of adaptive or robust discrete event systems in multiagent applications.

The control of discrete event systems

An overview of one trend among others in the development of a control theory for discrete-event systems and it can fairly be said that control of DES is now an established branch of control theory.

Planning and Control

This book is useful to researchers in artificial intelligence and control theory, and others concerned with the design of complex applications in robotics, automated manufacturing, and time-critical decision support.

Verifying an intelligent structural control system: a case study

Describes the formal verification of the timing properties of the design of an intelligent structural control system using the Concurrency Workbench, an automatic verification tool for finite-state

On Explicit Plan Languages for Coordinating Multiagent Plan Execution

An agent coordinating with another agent needs a model of the other agent to promote cooperation or to avoid interference. Agent plans in their executable form, however, are not well suited for being