New Zealand

oshiyuki Tajima and Takehisa Onisawa (February 1st 2010). Model-based Reinforcement Learning with Model Error and Its Application, Application of Machine Learning, Yagang Zhang, IntechOpen, DOI: 10.5772/8607.

Kunikazu Kobayashi, Koji Nakano, Takashi Kuremoto and Masanao Obayashi (February 1st 2010). Objective-based Reinforcement Learning System for Cooperative Behavior Acquisition, Application of Machine Learning, Yagang Zhang, IntechOpen, DOI: 10.5772/8615. 

Ilaria Giannoccaro and Pierpaolo Pontrandolfo (February 1st 2008). How Negotiation Influences the Effective Adoption of the Revenue Sharing Contract: A Multi-Agent Systems Approach, Supply Chain, Vedran Kordic, IntechOpen, DOI: 10.5772/5337.

Khair Eddin Sabri, Ridha Khedri and Jason Jaskolka (October 1st 2009). Algebraic Model for Agent Explicit Knowledge in Multi-agent Systems, Advanced Technologies, Kankesu Jayanthakumaran, IntechOpen, DOI: 10.5772/8211. 

In many multi-agent interactions in the real world, agents receive payoffs over multiple distinct criteria; i.e. the payoffs are multi-objective in nature. However, the same multi-objective payoff vector may lead to different utilities for each participant. Therefore, it is essential for an agent to learn about the behaviour of other agents in the system. We investigate here the effects of opponent modelling on multi-objective multi-agent interactions with non-linear utilities. Specifically, we consider multi-objective normal form games (MONFGs) with non-linear utility functions under the scalarised expected returns optimisation criterion. We contribute a novel actor-critic formulation to allow reinforcement learning of mixed strategies in this setting, along with an extension that incorporates opponent policy reconstruction using conditional action frequencies. Our empirical results demonstrate that opponent modelling can drastically alter the learning dynamics in this setting.


AAMAS 2020

Opponent Modelling for Reinforcement Learning in Multi-Objective Normal Form Games

opponent modelling

multi-objective decision making

multi-agent systems

reinforcement learning

AAMAS is the leading scientific conference for research in autonomous agents and multi-agent systems. The AAMAS conference series was initiated in 2002 as the merging of three respected scientific meetings: the International Conference on Multi-Agent Systems (ICMAS), the International Workshop on Agent Theories, Architectures, and Languages (ATAL), and the International Conference on Autonomous Agents (AA). The aim of the joint conference is to provide a single, high-profile, internationally-respected archival forum for scientific research in the theory and practice of autonomous agents and multi-agent systems.

Browse keynotes, discussions, panels and over 300 presentations.


AAMAS is the leading scientific conference for research in autonomous agents and multi-agent systems. The AAMAS conference series was initiated in 2002 as the merging of three respected scientific meetings: the International Conference on Multi-Agent Systems (ICMAS), the International Workshop on Agent Theories, Architectures, and Languages (ATAL), and the International Conference on Autonomous Agents (AA). 

technical paper

Too many coocks: Coordinating multi-agent collaboration through inverse planning

Reinforcement learning and planning have been revolutionized in recent years, due in part to the mass adoption of deep convolutional neural networks and the resurgence of powerful methods to refine decision-making policies. However, the problem of sparse reward signals remains pervasive in many domains. While various reward-shaping mechanisms and imitation learning approaches have been proposed to mitigate this problem, a mathematically rigorous structure of the underlying objective is rarely exploited. In this paper, we resolve this by defining objectives using linear temporal logic over finite traces (LTL_f) and utilize the automaton representation of said objectives in order to define novel reward shaping functions that mitigate the sparse rewards problem within modern Monte Carlo Tree Search (MCTS) methods. We further demonstrate that such verification-guided reward shaping can be utilized to facilitate transfer learning between different environments when the objective is the same.


Verification-Guided Tree Search

This presentation describes the use of CArtAgO artifacts to represent virtual resources exposed by an agent in a hypermedia context as an extension of the Multi-Agent MicroServices architecture.

Explicit Modelling of Resources for Multi-Agent Microservices using the CArtAgo framework

Simulated games have become a dominant platform for multiagent intelligence research in recent years. Previous works have succeeded on arcade, first person shooter (FPS), real-time strategy (RTS), and massive online battle arena (MOBA) games. Our work considers massively multiplayer online role-playing games (MMORPGs or MMOs), which capture several complexities of real-world learning that are not well modeled by any other game genre. We present a massively multiagent game environment inspired by MMOs and demonstrate that simple policy gradient methods produce interesting emergent exploration and specialization behaviors.


Neural MMO v1.3: A Massively Multiagent Game Environment for Training and Evaluating Neural Networks

A key assumption in Distributed Constraint Optimization Problem (DCOP) model is that all constraints are fully specified or known a priori, which may not hold in applications where constraints encode preferences of human users. We extend the model to Incom- plete DCOPs (I-DCOPs), where some constraints can be partially specified. User preferences for these partially-specified constraints can be elicited during the execution of I-DCOP algorithms, but they incur some elicitation costs. Additionally, we extend the Synchro- nous Branch-and-Bound (SyncBB) algorithm to solve I-DCOPs. Our model extends the state of the art in distributed constraint reason- ing to better model and solve distributed agent-based applications with user preferences.


Embedding Preference Elicitation Within the Search for DCOP Solutions

Distributed Constraint Optimization Problems (DCOPs) are a suitable formulation for coordinating interactions (i.e. constraints) in cooperative multi-agent systems. The traditional DCOP model deals with variables that can take only discrete values. However, there are many applications where the variables are continuous decision variables. The existing methods for solving DCOPs with continuous variables come with a huge computation and communication overhead. In this paper, we apply continuous non-linear optimization methods on Cooperative Constraint Approximation (CoCoA) algorithm. Empirical results show that our algorithm is able to provide high-quality solutions at the expense of small communication cost and execution time.


C-CoCoA: A Continuous Cooperative Constraint Approximation Algorithm to Solve Functional DCOPs

The paper introduces a generic approach to solving Sequential Security Games (SGs) which utilizes Evolutionary Algorithms (EAs). Formulation of the method (named EASG) is general and largely game-independent, which allows for its application to a wide range of SGs with just little adjustments addressing game specificity. Experiments performed on 3 different types of games (with 300 instances in total) demonstrate robustness and stability of EASG, manifested by repeatable achieving optimal or near-optimal solutions in the vast majority of the cases. The main advantage of EASG is time efficiency. The method scales better than state-of-the-art approaches and can be applied to sequential SGs with bigger numbers of steps compared to the existing methods. Due to anytime characteristics, EASG is very well suited for time-critical applications.


A Generic Metaheuristic Approach to Sequential Security Games

This paper considers the fairness in the problem of budget-feasible mechanism design in two-sided markets where multiple sellers come with indivisible items and buyers come with budgets. Buyers could untruthfully claim their budgets to procure as much value of items as possible from sellers. Each seller with a single item is required to bid his cost since the cost is privately known, while the value of each item is publicly known. A viable mechanism should satisfy buyers’ fairness where a buyer with more budget can procure more value of items, and budget feasibility where buyers’ respective budgets are not exceeded. The goal is to investigate budget-feasible mechanisms that guarantee the fairness, incentives and efficiency simultaneously. We consider two models by distinguishing the types of items, one with homogeneous items and one with heterogeneous items. Our main contributions are the budget-feasible mechanisms for these models that guarantee the fairness, the truthfulness both on the sellers’ side and the buyers’ side, and constant approximation to the optimal total procured value from sellers.


Two-sided Auctions with Budgets: Fairness, Incentives and Efficiency

We develop a Multi-Agent Reinforcement Learning (MARL) method that finds approximately optimal policies for cooperative agents that co-exist in an environment. Central to achieving this is how the agents learn to communicate with each other. Can they together develop a language while learning to perform a common task? We formulate and study a MARL problem where cooperative agents are connected via a fixed underlying network. These agents communicate along the edges of this network by exchanging discrete symbols. However, the semantics of these symbols are not predefined and have to be learned during the training process. We propose a method for training these agents using emergent communication. We demonstrate the applicability of the proposed framework by applying it to the problem of managing traffic controllers, where we achieve state-of-the-art performance (as compared to several strong baselines) and perform a detailed analysis of the emergent communication.


Networked Multi-Agent Reinforcement Learning with Emergent Communication

Humans use language to collectively execute abstract strategies besides using it as a referential tool for identifying physical entities. In our paper, we study the role of emergent languages in discovering and implementing strategies. We formulate the problem using a voting game where two candidate agents contest in an election with the goal of convincing population members (other agents), that are connected to each other via an underlying network, to vote for them. To achieve this goal, agents are only allowed to exchange messages in the form of sequences of discrete symbols. Using our proposed framework, we answer the following questions: (i) Do the agents learn to communicate in a meaningful way? (ii) Does the system evolve as expected under various reward structures? (iii) How is the emergent language affected by the community structure in the network? To the best of our knowledge, we are the first to study emergence of communication among networked agents for discovering and implementing strategies.

Downloads

Next from AAMAS 2020

Too many coocks: Coordinating multi-agent collaboration through inverse planning

Similar lecture

Opponent Modelling using Policy Reconstruction for Multi-Objective Normal Form Games

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAMAS 2020

Too many coocks: Coordinating multi-agent collaboration through inverse planning

Similar lecture

Opponent Modelling using Policy Reconstruction for Multi-Objective Normal Form Games

Downloads