Interaction and cooperation among agents are two fundamental activities having place in multi-agent systems. In this regard, in a generally large universe of agents, the main difficulty lies in determining which agents are the most promising candidates for achieving fruitful interaction (i.e., cooperation). In a holistic view, several strategies like trust, reliability, reputation, cyphering or commitment can play, for example, a crucial role in various online environments. They provide a multitude of mechanisms for assessing credibility, trustworthiness, identity and so on, not only with respect to agents but also considering entities such as users, products or services. Agent-based scenarios can take advantage from these techniques, which are also able to reduce information asymmetry between agents, enhancing their mutual confidence and, in this way, encouraging their participation by limiting the probability of deception. Moreover, such systems form the basis for a potentially large number of countermeasures to prevent, depotentiate or nullify malicious activities intended at altering a correct perception about the counterpart by means of misrepresentation or manipulation activities carried out by bad agents. In this context, the goal of this chapter is to provide an overview of the several specific proposals, made at Workshop on Objects and Agents (WOA) over more than two decades, aimed at identifying the best partners for interaction and cooperation based on the above-cited approaches.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Looking for the Best Partners

  • Domenico Rosaci,
  • Giuseppe M. L. Sarnè

摘要

Interaction and cooperation among agents are two fundamental activities having place in multi-agent systems. In this regard, in a generally large universe of agents, the main difficulty lies in determining which agents are the most promising candidates for achieving fruitful interaction (i.e., cooperation). In a holistic view, several strategies like trust, reliability, reputation, cyphering or commitment can play, for example, a crucial role in various online environments. They provide a multitude of mechanisms for assessing credibility, trustworthiness, identity and so on, not only with respect to agents but also considering entities such as users, products or services. Agent-based scenarios can take advantage from these techniques, which are also able to reduce information asymmetry between agents, enhancing their mutual confidence and, in this way, encouraging their participation by limiting the probability of deception. Moreover, such systems form the basis for a potentially large number of countermeasures to prevent, depotentiate or nullify malicious activities intended at altering a correct perception about the counterpart by means of misrepresentation or manipulation activities carried out by bad agents. In this context, the goal of this chapter is to provide an overview of the several specific proposals, made at Workshop on Objects and Agents (WOA) over more than two decades, aimed at identifying the best partners for interaction and cooperation based on the above-cited approaches.