A Machine Learning Approach for Mechanism Selection in Complex Negotiations

Abstract

Automated negotiation mechanisms can be helpful in contexts where users want to reach mutually satisfactory agreements about issues of shared interest, especially for complex problems with many interdependent issues. A variety of automated negotiation mechanisms have been proposed in the literature. The effectiveness of those mechanisms, however, may depend on the characteristics of the underlying negotiation problem (e.g. on the complexity of participant's utility functions, as well as the degree of conflict between participants). While one mechanism may be a good choice for a negotiation problem, it may be a poor choice for another. In this paper, we pursue the problem of selecting the most effective negotiation mechanism given a particular problem by (1) defining a set of scenario metrics to capture the relevant features of negotiation problems, (2) evaluating the performance of a range of negotiation mechanisms on a diverse test suite of negotiation scenarios, (3) applying machine learning techniques to identify which mechanisms work best with which scenarios, and (4) demonstrating that using these classification rules for mechanism selection enables significantly better negotiation performance than any single mechanism alone.

Decision-Making under Uncertainty for the Deployment of Future Hyperconnected Networks: A Survey

“…• Enhanced decisions. These studies focus on improving the performance of decisionmaking results by choosing selection mechanisms in complex negotiation scenarios [102], by considering environmental awareness when dealing with critical complex systems processes under supervision [103], by assessing the advantages of combining the learning process with multiple agents and weighted strategies [104], and by suggesting two complementary stages that would exist between machine learning and support vector machines [105]. • Complex systems.…”

Expectation: Personalized Explainable Artificial Intelligence for Decentralized Agents with Heterogeneous Knowledge

“…AT are mostly based on heuristics [24,25] and traditional ML methods (e.g., decision trees [26,27], Bayesian learning [28,29,30], and concept-based learning [31,32]) and rely on possibly numerous bid exchanges regulated by negotiation protocols [33]. By exploiting such techniques, machines can negotiate with humans seamlessly, resolving conflicts with a high degree of mutual understanding [34].…”

A Data-Driven Method for Recognizing Automated Negotiation Strategies

“…We consider MyAgent as experienced in a domain if it has negotiated in that domain (for real or in simulations), so that D includes negotiation traces from that domain. We consider MyAgent's knowledge of OpAgent's preference as (1) complete, if MyAgent knows the exact preference profile of OpAgent (which can be the case in some repeated negotiations); (2) partial, if MyAgent has encountered OpAgent in the past but the uncertainty of the estimated opponent profile is low or if the domain is partially predictable [2]; and (3) none, if MyAgent has not negotiated with the opponent before and the domain is not predictable.…”