{"title":"COMB: Scalable Concession-Driven Opponent Models Using Bayesian Learning for Preference Learning in Bilateral Multi-Issue Automated Negotiation","authors":"Shengbo Chang, Katsuhide Fujita","doi":"10.1007/s10726-024-09889-7","DOIUrl":null,"url":null,"abstract":"<p>Learning an opponent’s preferences in bilateral multi-issue automated negotiations can lead to more favorable outcomes. However, existing opponent models can fail in negotiation contexts when their assumptions about opponent behaviors differ from actual behavior patterns. Although integrating broader behavioral assumptions into these models could be beneficial, it poses a challenge because the models are designed with specific assumptions. Therefore, this study proposes an adaptable opponent model that integrates a general behavioral assumption. Specifically, the proposed model uses Bayesian learning (BL), which can apply various behavioral assumptions by considering the opponent’s entire bidding sequence. However, this BL model is computationally infeasible for multi-issue negotiations. Hence, current BL models often impose constraints on their hypothesis space, but these constraints about the utility function’s shape significantly sacrifice accuracy. This study presents a novel scalable BL model that relaxes these constraints to improve accuracy while maintaining linear time complexity by separately learning each parameter of a utility function. Furthermore, we introduce a general assumption that the opponent’s bidding strategy follows a concession-based pattern to enhance adaptability to various negotiation contexts. We explore three likelihood function options to implement this assumption effectively. By incorporating these options into the proposed scalable model, we develop three scalable concession-driven opponent models using Bayesian learning (COMB). Experiments across 45 negotiation domains using 15 basic agents and 15 finalists from the automated negotiating agents competition demonstrate the proposed scalable model’s higher accuracy than existing scalable models. COMB models show higher adaptability to various negotiation contexts than state-of-the-art models.</p>","PeriodicalId":47553,"journal":{"name":"Group Decision and Negotiation","volume":"16 1","pages":""},"PeriodicalIF":3.6000,"publicationDate":"2024-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Group Decision and Negotiation","FirstCategoryId":"91","ListUrlMain":"https://doi.org/10.1007/s10726-024-09889-7","RegionNum":4,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MANAGEMENT","Score":null,"Total":0}
引用次数: 0
Abstract
Learning an opponent’s preferences in bilateral multi-issue automated negotiations can lead to more favorable outcomes. However, existing opponent models can fail in negotiation contexts when their assumptions about opponent behaviors differ from actual behavior patterns. Although integrating broader behavioral assumptions into these models could be beneficial, it poses a challenge because the models are designed with specific assumptions. Therefore, this study proposes an adaptable opponent model that integrates a general behavioral assumption. Specifically, the proposed model uses Bayesian learning (BL), which can apply various behavioral assumptions by considering the opponent’s entire bidding sequence. However, this BL model is computationally infeasible for multi-issue negotiations. Hence, current BL models often impose constraints on their hypothesis space, but these constraints about the utility function’s shape significantly sacrifice accuracy. This study presents a novel scalable BL model that relaxes these constraints to improve accuracy while maintaining linear time complexity by separately learning each parameter of a utility function. Furthermore, we introduce a general assumption that the opponent’s bidding strategy follows a concession-based pattern to enhance adaptability to various negotiation contexts. We explore three likelihood function options to implement this assumption effectively. By incorporating these options into the proposed scalable model, we develop three scalable concession-driven opponent models using Bayesian learning (COMB). Experiments across 45 negotiation domains using 15 basic agents and 15 finalists from the automated negotiating agents competition demonstrate the proposed scalable model’s higher accuracy than existing scalable models. COMB models show higher adaptability to various negotiation contexts than state-of-the-art models.
期刊介绍:
The idea underlying the journal, Group Decision and Negotiation, emerges from evolving, unifying approaches to group decision and negotiation processes. These processes are complex and self-organizing involving multiplayer, multicriteria, ill-structured, evolving, dynamic problems. Approaches include (1) computer group decision and negotiation support systems (GDNSS), (2) artificial intelligence and management science, (3) applied game theory, experiment and social choice, and (4) cognitive/behavioral sciences in group decision and negotiation. A number of research studies combine two or more of these fields. The journal provides a publication vehicle for theoretical and empirical research, and real-world applications and case studies. In defining the domain of group decision and negotiation, the term `group'' is interpreted to comprise all multiplayer contexts. Thus, organizational decision support systems providing organization-wide support are included. Group decision and negotiation refers to the whole process or flow of activities relevant to group decision and negotiation, not only to the final choice itself, e.g. scanning, communication and information sharing, problem definition (representation) and evolution, alternative generation and social-emotional interaction. Descriptive, normative and design viewpoints are of interest. Thus, Group Decision and Negotiation deals broadly with relation and coordination in group processes. Areas of application include intraorganizational coordination (as in operations management and integrated design, production, finance, marketing and distribution, e.g. as in new products and global coordination), computer supported collaborative work, labor-management negotiations, interorganizational negotiations, (business, government and nonprofits -- e.g. joint ventures), international (intercultural) negotiations, environmental negotiations, etc. The journal also covers developments of software f or group decision and negotiation.