NICOP - Distributed Symbolic-Non-Symbolic Context-Aware Swarm Logics
Abstract
Distributed decision-making is a necessity in most large-scale real-world contexts, where theenvironment does not afford the agents unlimited resources, communication capabilitiesand/or time to act. Representing and learning situation and task specific contexts afford eachdecision node the ability to tailor critical choices made during action production andreasoning to the constraints and objectives defined by these contexts. The coupling betweenrepresentation and learning of contexts on the one hand and action production and reasoningon the other hand entails that the overall ???machine cognition??? architecture needs to span bothperception and planning functions.The primary technical objective of the proposed effort is to design new algorithms that blendthe advantages of symbolic and non-symbolic AI algorithms into swarm logics capable ofrepresenting and learning contexts successfully and efficiently in a distributed environment.This project will make a leap forward in distributed artificial intelligence by creating efficientrepresentations for contexts, fast, memory efficient, and transparent context learningalgorithms, and symbolic and non-symbolic planning tools for swarm logics.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 04, 2018
- Source ID
- N629091812140
Entities
People
- Hussein A. Abbass
Organizations
- Office of Naval Research
- United States Navy
- University of New South Wales