Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MiCRO for Multilateral Negotiations

David Aguilera-Luzon, Dave de Jonge, Javier Larrosa

TL;DR

This paper extends the MiCRO negotiation strategy from bilateral to multilateral settings by introducing MiCRO-Multi and evaluating it against top ANAC multilateral agents (2015, 2017, 2018). Using NegMAS with Genius bridge, it demonstrates that MiCRO-Multi achieves high mean utility and forms empirical Nash equilibria under best-response dynamics, despite lacking opponent modeling or parameter tuning. The results show MiCRO remains competitive across diverse multilateral domains and highlight that traditional ANAC benchmarks may be insufficiently challenging, potentially being exploitable by simple, reactive strategies. The work argues for richer, more adversarial benchmarking environments and outlines directions for domain complexity analysis, robustness testing, and improved evaluation platforms to better distinguish strategic sophistication.

Abstract

Recently, a very simple new bilateral negotiation strategy called MiCRO was introduced that does not make use of any kind of opponent modeling or machine learning techniques and that does not require fine-tuning of any parameters. Despite its simplicity, it was shown that MiCRO performs similar to -- or even better than -- most state-of-the-art negotiation strategies. This lead its authors to argue that the benchmark domains on which negotiation algorithms are typically tested may be too simplistic. However, one question that was left open, was how MiCRO could be generalized to multilateral negotiations. In this paper we fill this gap by introducing a multilateral variant of MiCRO. We compare it with the winners of the Automated Negotiating Agents Competitions (ANAC) of 2015, 2017 and 2018 and show that it outperforms them. Furthermore, we perform an empirical game-theoretical analysis to show that our new version of MiCRO forms an empirical Nash equilibrium.

MiCRO for Multilateral Negotiations

TL;DR

This paper extends the MiCRO negotiation strategy from bilateral to multilateral settings by introducing MiCRO-Multi and evaluating it against top ANAC multilateral agents (2015, 2017, 2018). Using NegMAS with Genius bridge, it demonstrates that MiCRO-Multi achieves high mean utility and forms empirical Nash equilibria under best-response dynamics, despite lacking opponent modeling or parameter tuning. The results show MiCRO remains competitive across diverse multilateral domains and highlight that traditional ANAC benchmarks may be insufficiently challenging, potentially being exploitable by simple, reactive strategies. The work argues for richer, more adversarial benchmarking environments and outlines directions for domain complexity analysis, robustness testing, and improved evaluation platforms to better distinguish strategic sophistication.

Abstract

Recently, a very simple new bilateral negotiation strategy called MiCRO was introduced that does not make use of any kind of opponent modeling or machine learning techniques and that does not require fine-tuning of any parameters. Despite its simplicity, it was shown that MiCRO performs similar to -- or even better than -- most state-of-the-art negotiation strategies. This lead its authors to argue that the benchmark domains on which negotiation algorithms are typically tested may be too simplistic. However, one question that was left open, was how MiCRO could be generalized to multilateral negotiations. In this paper we fill this gap by introducing a multilateral variant of MiCRO. We compare it with the winners of the Automated Negotiating Agents Competitions (ANAC) of 2015, 2017 and 2018 and show that it outperforms them. Furthermore, we perform an empirical game-theoretical analysis to show that our new version of MiCRO forms an empirical Nash equilibrium.
Paper Structure (19 sections, 5 equations, 1 figure, 2 tables)

This paper contains 19 sections, 5 equations, 1 figure, 2 tables.

Table of Contents

  1. Introduction
  2. Benchmarking Through ANAC
  3. Scope and Objectives
  4. Formalization of Core Concepts in Automated Negotiation
  5. ANAC Scenario Configuration (2015--2018)
  6. The MiCRO Benchmark Strategy
  7. How MiCRO Works
  8. New MiCRO Strategy for Multi-lateral Scenarios
  9. The Deadlock Problem.
  10. Experimental Setup
  11. Selection of Opponent Agents
  12. Negotiation Scenarios
  13. Test Methodologies
  14. Performance Metrics
  15. Results
  16. ...and 4 more sections

Figures (1)

  • Figure 1: Best-response dynamics across all triplets of strategy profiles. Arrows indicate unilateral improvements; Nash equilibria are highlighted.