Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Merging plans with incomplete knowledge about actions and goals through an agent-based reputation system

Javier Carbo, Jose M Molina, Miguel A Patricio

TL;DR

This work tackles the challenge of generating transition plans for individuals with cognitive disabilities by framing it as a distributed, multi-agent planning problem with incomplete knowledge about actions and goals. It introduces a domain-independent agent-based framework in which operator and node agents exchange recommendations and use a contextual reputation system to merge plans. Four merging strategies are proposed, ranging from no-merge baselines to reputation-guided combination and crossover, all evaluated with ART-inspired performance measures. The findings suggest that leveraging recommendations yields greater improvements than the specific merging method itself, and the authors provide an open-source JADEX framework to enable further research and practical deployment for autism-related planning assistance.

Abstract

Managing transition plans is one of the major problems of people with cognitive disabilities. Therefore, finding an automated way to generate such plans would be a helpful tool for this community. In this paper we have specifically proposed and compared different alternative ways to merge plans formed by sequences of actions of unknown similarities between goals and actions executed by several operator agents which cooperate between them applying such actions over some passive elements (node agents) that require additional executions of another plan after some time of use. Such ignorance of the similarities between plan actions and goals would justify the use of a distributed recommendation system that would provide an useful plan to be applied for a certain goal to a given operator agent, generated from the known results of previous executions of different plans by other operator agents. Here we provide the general framework of execution (agent system), and the different merging algorithms applied to this problem. The proposed agent system would act as an useful cognitive assistant for people with intelectual disabilities such as autism.

Merging plans with incomplete knowledge about actions and goals through an agent-based reputation system

TL;DR

This work tackles the challenge of generating transition plans for individuals with cognitive disabilities by framing it as a distributed, multi-agent planning problem with incomplete knowledge about actions and goals. It introduces a domain-independent agent-based framework in which operator and node agents exchange recommendations and use a contextual reputation system to merge plans. Four merging strategies are proposed, ranging from no-merge baselines to reputation-guided combination and crossover, all evaluated with ART-inspired performance measures. The findings suggest that leveraging recommendations yields greater improvements than the specific merging method itself, and the authors provide an open-source JADEX framework to enable further research and practical deployment for autism-related planning assistance.

Abstract

Managing transition plans is one of the major problems of people with cognitive disabilities. Therefore, finding an automated way to generate such plans would be a helpful tool for this community. In this paper we have specifically proposed and compared different alternative ways to merge plans formed by sequences of actions of unknown similarities between goals and actions executed by several operator agents which cooperate between them applying such actions over some passive elements (node agents) that require additional executions of another plan after some time of use. Such ignorance of the similarities between plan actions and goals would justify the use of a distributed recommendation system that would provide an useful plan to be applied for a certain goal to a given operator agent, generated from the known results of previous executions of different plans by other operator agents. Here we provide the general framework of execution (agent system), and the different merging algorithms applied to this problem. The proposed agent system would act as an useful cognitive assistant for people with intelectual disabilities such as autism.
Paper Structure (12 sections, 10 equations, 9 figures)

This paper contains 12 sections, 10 equations, 9 figures.

Table of Contents

  1. Introduction
  2. General context and motivation
  3. Hold assumptions
  4. Defining the agent system
  5. Merging Plans
  6. Computing the reputation of an operator agent and the required time for the next plan execution
  7. Merging method 0: no merging ignoring recommendations
  8. Merging method 1: no merging considering recommendations
  9. Merging method 2
  10. Merging method 3
  11. Experimentation
  12. Conclusions

Figures (9)

  • Figure 1: CFP interaction protocol between a node and operator agent
  • Figure 2: Propose interaction protocol between a node and operator agent
  • Figure 3: Query recommenders interaction protocol between two operator agents
  • Figure 4: Query expertise interaction protocol between two operator agents
  • Figure 5: Query recommendation interaction protocol between two operator agents
  • ...and 4 more figures