Selective Exploration Algorithm for Coalition Formation

This work focuses on coalition formation among heterogeneous agents for the 2017 multiagent programming contest. An agent is a computer system that is capable of independent action to achieve its goals. In order to increase the effectiveness of the agents, we can organise them into coalitions, in which the agents collaborate with each other to achieve individual or common goals. We investigate and apply coalition structure generation (the first activity of the coalition formation process) in simulated scenarios, specifically the 2017 contest scenario, where the agents forming a competing team cooperate to solve logistic problems simulated on the map of a real city. In order to achieve our goal, we integrate coalition formation algorithms into the JaCaMo platform by means of a CArtAgO artefact, named CFArtefact. We use the implementation of the SMART JaCaMo team for experimenting with the coalition formation approach in the contest scenario. We experiment on three approaches in the contest domain with different configurations. In the first, we use only a taskallocation mechanism, while the other approaches use an optimal coalition formation algorithm and a heuristic coalition formation algorithm. We conducted several experiments to compare the advantages of each approach. Our results show that coalition formation algorithms can improve the performance of a participating team when dealing with low job rates (i.e., how quickly new jobs are created by the simulation). However, as we increase the job rate, the approach using only task allocation has better performance. Even a heuristic coalition formation approach has close performance to the optimal one in that case. Coalition formation can play an important role when we aim to balance each group of agents to accomplish some particular goal given a larger team of cooperating agents.

We present a simulated annealing algorithm for coalition formation represented as characteristic function games. We provide a detailed analysis of various neighbourhoods for the algorithm, and of their effects to the algorithm’s performance. Our practical experiments and comparisons with other methods demonstrate that simulated annealing provides a useful tool to tackle the combinatorics involved in multi-agent coalition formation.

Incorporating coalition formation algorithms into agent systems shall be advantageous due to the consequent increase in the overall quality of task performance. Coalition formation was addressed in game theory, however the game theoretic approach is centralized and computationally intractable. Recent work in DAI has resulted in distributed algorithms with computational tractability. This paper addresses the implementation of distributed coalition formation algorithms within a real-world multi-agent system. We present the problems that arise when attempting to utilize the theoretical coalition formation algorithms for a real-world system, demonstrate how some of their restrictive assumptions can be relaxed, and discuss the resulting benefits. In addition, we analyze the modifications, the complexity and the quality of the cooperation mechanisms. The task domain of our multi-agent system is information gathering, filtering and decision support within the WWW.

This paper presents two bio-inspired algorithms for coalition formation of multiple modular robot systems. An effective and efficient coalition formation system can help modular robot system take full advantage of reconfigurability of modular robots. In this paper, the multirobot coalition formation problem is illustrated and a mathematical model for the problem is described. Two bio-inspired algorithms, ant-colony algorithm (ACA) and genetic algorithm (GA), are introduced for solving the mathematical model. With the two algorithms, it is able to form a large number of robots into many different groups for a variety of applications, such as parallel performance of multiple tasks by multiple teams of robots. The paper compares the efficiency and effectiveness of two algorithms for solving the presented problem with case study. The results for the comparison study are analyzed and discussed. Also, the implementation details of the simulation and experiment using ACA are presented in the paper.

Recently there have been numerous studies exploring the benefits of the coalition formation in a cognitive radio network and it has been shown that coalition formation tends to improve the performance of cognitive radio networks. In this paper, we use the concepts from matching theory to form coalitions of varying size among cognitive radio users for cooperative spectrum sensing under target detection probability constraint. In the proposed model, we make use of one-sided matching theory to develop an algorithm for the cognitive radios to form coalitions of varying the size to improve their individual gains (.e.g. throughput and the probability of false alarm). We prove that the proposed algorithm leads to stable coalition formation in the cognitive radio network and show using simulations that the proposed matching algorithm for coalition formation yields significant gains in term of reduced false alarm probability and increased throughput per cognitive radio user as compared to the non-cooperative scenario.

SummaryWe consider fog computing scenarios where data generated by a set of IoT applications need to be processed locally by a set of fog nodes, belonging to distinct Fog Infrastructure Providers (FIPs) sharing the same co‐location facility, with the aim of increasing their profits. This is a challenging goal as it requires reducing costs and meeting QoS targets despite time‐varying workloads. We argue that these FIPs may find it profitable to cooperate by mutually sharing their workload and resources, and we show (by using a game‐theoretical framework) that this is indeed the case when stable coalitions can be formed. Based on these results, in this paper, we present (1) a mathematical model for maximizing the profit obtained for allocating IoT applications to a group of FIPs, and (2) a coalition formation algorithm that allows each FIP to decide with whom to cooperate so as to increment its profits. The efficacy of the devised algorithm is assessed by means of an experimental evaluation taking into account different workload intensities. The results from these experiments show the capability of the proposed algorithm to form coalitions of FIPs that are profitable and stable in all the scenarios we take into consideration.

We consider a cognitive radio setting in which a set of secondary transmitter-receiver pairs coexist with a primary user. In the channel sensing phase, each secondary system senses the primary user's channel to determine its occupancy. The existence of a fusion center enables joint processing of the sensing data. This would improve the joint sensing performance, however at the cost of a delay required for reporting the sensed data. In the channel access phase, noncooperative secondary transmission can lead to a degradation of secondary users' average achievable rates due to excessive mutual interference. We propose to coordinate the secondary system transmission by providing orthogonal multiple access rules for the secondary users which report their sensing data to the fusion center. Using models from game theory, a coalition formation algorithm is proposed to find a stable set of secondary users participating in sensing data fusion. It is shown by simulation that the proposed mechanism provides an incentive for the secondary users to participate in joint sensing at an extent depending on the signal-to-noise ratio and the reporting overhead.

In this paper, the problem of random access contention between machine type devices (MTDs) in the uplink of a wireless cellular network is studied. In particular, the possibility of forming cooperative groups to coordinate the MTDs’ requests for the random access channel (RACH) is analyzed. The problem is formulated as a stochastic coalition formation game in which the MTDs are the players that seek to form cooperative coalitions to optimize a utility function that captures each MTD’s energy consumption and time-varying queue length. Within each coalition, an MTD acts as a coalition head that sends the access requests of the coalition members over the RACH. One key feature of this game is its ability to cope with stochastic environments in which the arrival requests of MTDs and the packet success rate over RACH are dynamically time-varying. The proposed stochastic coalitional game is composed of multiple stages, each of which corresponds to a coalitional game in stochastic characteristic form that is played by the MTDs at each time step. To solve this game, a novel distributed coalition formation algorithm is proposed and shown to converge to a stable MTD partition. Simulation results show that, on the average, the proposed stochastic coalition formation algorithm can reduce the average fail ratio and energy consumption of up to 36% and 31% for a cluster-based distribution of MTDs, respectively, compared with a noncooperative case. Moreover, when the MTDs are more sensitive to the energy consumption (queue length), the coalitions’ size will increase (decrease).

Coalition formation with dynamically changing externalities

Decentralized, partially or fully distributed multi-agent coordination, cooperation and coalition formation have been studied extensively over the past 20 years. Ensembles of autonomous software, robotic or other autonomous agents may need to coordinate and form coalitions in order to share resources, divide-and-conquer tasks, or enable themselves to complete tasks too complex for any individual agent alone. Much of prior research on multi-agent coordination and coalition formation focuses on the distributed problem solving context, in which all agents belong to the same end-user and share the overarching system-level objectives. As the Internet is evolving towards “Web 3.0” and “the Internet of Things”, modeling systems and designing protocols for scenarios in which autonomous agents that belong to different end-users, and hence in general do not share common goals, yet may still need to coordinate and cooperate with each other, is becoming a research challenge of paramount importance. In this paper, we outline a general vision and some key properties of an “Internet of (Software) Agents”, and propose a reputation-based model for effective multi-agent coordination among autonomous agents that need to coordinate with (yet don’t a priori know to what extent they can trust) each other.

Hedonic coalition formation for optimal deployment

Target tracking in stand-forward swarm jamming for unmanned aerial vehicles (UAVs) is a promising topic but the relevant research has been tepid. This paper aims at exploring this problem for multistatic radar-enabled UAVs (RUAVs) via multiagent cooperative sensing and control scheme, in which all RUAVs constitute a multistatic radar system with variable and movable transmitting and receiving platforms, and each RUAV as an autonomous agent makes decision with other agents for transmit-receive task assignment and path planning to seek the optimal overall target tracking performance. The optimization problem of RUAVs is formulated as minimizing the Bayesian Cramer-Rao lower bound with some practical constraints. ´ To solve this problem, an alternating iterative optimization scheme is presented, in which the coalition formation method is applied for transmit-receive task assignment and the improved particle swarm optimization (PSO) method is devised for path planning. The alternating coalition formation and PSO (ACFPSO) algorithm is therefore proposed for RUAVs for target tracking in stand-forward jamming. Numerical simulations in the typical scenario are conducted to verify the effectiveness and superiority of the ACFPSO algorithm from aspects of tracking accuracy and jamming elimination.

This paper considers the cooperative optimization of mutual information in the MIMO Gaussian interference channel in a fully distributed manner via game theory. Null shaping constraints are enforced in the design of transmit covariance matrices to enable interference mitigation among links. The transmit covariance matrices leading to the Nash Equilibrium (NE) are derived, and the existence and uniqueness of the NE is analyzed. The formation of the cooperative sets, that represent the cooperation relationship among links, is considered as coalition games and network formation games. We prove that the proposed coalition formation (CF) and coalition graph formation (CGF) algorithms are Nash-stable, and the proposed network formation (NF) algorithm converges to a Nash Equilibrium. Simulation results show that the proposed CF and CGF algorithms have significant advantages when the antennas at the transmitters is large, and the proposed NF algorithm enhances the sum rate of the system apparently even at low signal-to-noise ratio region and/or with small number of transmit antennas.

An important research problem in artificial intelligence is how to organize multiple agents, and coordinate them, so that they can work together to solve problems. Coordinating agents in a multi-agent system can significantly affect the system’s performance—the agents can, in many instances, be organized so that they can solve tasks more efficiently, and consequently benefit collectively and individually. Central to this endeavor is coalition formation—the process by which heterogeneous agents organize and form disjoint groups (coalitions). Coalition formation often involves finding a coalition structure (an exhaustive set of disjoint coalitions) that maximizes the system’s potential performance (e.g., social welfare) through coalition structure generation. However, coalition structure generation typically has no notion of goals. In cooperative settings, where coordination of multiple coalitions is important, this may generate suboptimal teams for achieving and accomplishing the tasks and goals at hand. With this in mind, we consider simultaneously generating coalitions of agents and assigning the coalitions to independent alternatives (e.g., tasks/goals), and present an anytime algorithm for the simultaneous coalition structure generation and assignment problem. This combinatorial optimization problem has many real-world applications, including forming goal-oriented teams. To evaluate the presented algorithm’s performance, we present five methods for synthetic problem set generation, and benchmark the algorithm against the industry-grade solver CPLEX using randomized data sets of varying distribution and complexity. To test its anytime-performance, we compare the quality of its interim solutions against those generated by a greedy algorithm and pure random search. Finally, we also apply the algorithm to solve the problem of assigning agents to regions in a major commercial strategy game, and show that it can be used in game-playing to coordinate smaller sets of agents in real-time.

Coalition formation over graphs is a well studied class of games whose players are vertices and feasible coalitions must be connected subgraphs. In this setting, the existence and computation of equilibria, under various notions of stability, has attracted a lot of attention. However, the natural process by which players, starting from any feasible state, strive to reach an equilibrium after a series of unilateral improving deviations, has been less studied. We investigate the convergence of dynamics towards individually stable outcomes under the following perspective: what are the most general classes of preferences and graph topologies guaranteeing convergence? To this aim, on the one hand, we cover a hierarchy of preferences, ranging from the most general to a subcase of additively separable preferences, including individually rational and monotone cases. On the other hand, given that convergence may fail in graphs admitting a cycle even in our most restrictive preference class, we analyze acyclic graph topologies such as trees, paths, and stars.