United Kingdom

High-performing teams learn effective communication strategies to judiciously share information and reduce the cost of communication overhead. Within multi-agent reinforcement learning, synthesizing effective policies requires reasoning about when to communicate, whom to communicate with, and how to process messages. We propose a novel multi-agent reinforcement learning algorithm, Multi-Agent Graph-attentIon Communication (MAGIC), with a graph-attention communication protocol in which we learn 1) a Scheduler to help with the problems of when to communicate and whom to address messages to, and 2) a Message Processor using Graph Attention Networks (GATs) with dynamic graphs to deal with communication signals. The Scheduler consists of a graph attention encoder and a differentiable attention mechanism, which outputs dynamic, differentiable graphs to the Message Processor, which enables the Scheduler and Message Processor to be trained end-to-end. We evaluate our approach on a variety of cooperative tasks, including Google Research Football. Our method outperforms baselines across all domains, achieving ~10.5% increase in reward in the most challenging domain. We also show MAGIC communicates $27.4%$ more efficiently on average than baselines, is robust to stochasticity, and scales to larger state-action spaces. Finally, we demonstrate MAGIC on a physical, multi-robot testbed.

AAMAS 2021

Multi-Agent Graph-Attention Communication and Teaming

AAMAS is the largest and most influential conference in the area of agents and multiagent systems, bringing together researchers and practitioners in all areas of agent technology and providing and internationally renowned high-profile forum for publishing and finding out about the latest developments in the field. AAMAS is the flagship conference of the non-profit International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS).  

![](https://assets.underline.io/uploads/markdown_image/1/image/ce4604bd54468a006164d16e080204fb.jpg)

Please click on the button below to register for this event.

If you are registered and see this page please check the account email you are logged in with. It must be the same as for the registration.

If you still have trouble accessing the contact please contact support at aamas2021@underline.io 

technical paper

Policy gradient methods have become one of the most popular classes of algorithms for multi-agent reinforcement learning. A key challenge, however, that is not addressed by many of these methods is multi-agent credit assignment: assessing an agent's contribution to the overall performance, which is crucial for learning good policies. We propose a novel algorithm called Dr.Reinforce that explicitly tackles this by combining difference rewards with policy gradients to allow for learning decentralized policies when the reward function is known. By differencing the reward function directly, Dr.Reinforce avoids difficulties associated with learning the 
-function as done by Counterfactual Multiagent Policy Gradients (COMA), a state-of-the-art difference rewards method. For applications where the reward function is unknown, we show the effectiveness of a version of Dr.Reinforce that learns a reward network that is used to estimate the difference rewards.

Difference Rewards Policy Gradients

In this work we introduce an interactive machine teaching approach that teaches a classification task to the learner. Our adaptive approach - Feature Importance Teaching (FIT) - does not assume perfect knowledge about the learner, as most machine teaching approaches do. It chooses, online, which sample to show next, as it updates the learner's model based on feedback from the student on the weights attributed to the features. We present simulated results where the student has a different prior knowledge from the one assumed by the teacher. The results have shown that our teaching approach can mitigate this mismatch and lead to a significantly faster learning curve than the ones obtained in conditions where the teacher randomly selects the samples or does not consider this feedback from the student.

Teaching Unknown Learners to Classify via Feature Importance

Recently, there is growing attention on applying deep reinforcement learning (DRL) to solve the 3D bin packing problem (3D BPP), given its favorable generalization and independence of ground-truth label. However, due to the relatively less informative yet computationally heavy encoder, and considerably large action space inherent to the 3D BPP, existing methods are only able to handle up to 50 boxes. In this paper, we propose to alleviate this issue via an end-to-end multimodal DRL agent, which sequentially addresses three sub-tasks of sequence, orientation and position, respectively. The resulting architecture enables the agent to solve large-scale instances of 100 boxes or more. Experiments show that the agent could learn highly efficient policies that deliver superior performance against all the baselines on instances of various scales.

Solving 3D Bin Packing Problem via Multimodal Deep Reinforcement Learning

This paper is concerned with designing purpose in autonomous robots for open-ended learning settings. Unconstrained human robot interaction situations and robotic systems that must operate in dynamic multi-robot scenarios are paradigmatic examples of open-endedness. An approach to the appropriate design and engineering of motivational structures to endow robots with a particular purpose is proposed and tested. This approach focuses on the drive structure and how it can be made to autonomously adapt to changing circumstances. Specifically, a simple evolutionary strategy for the autonomous regulation of multiple drives in order to optimize long-term operation is defined. The experimental results have been obtained on a Baxter robot facing changing situations in real setups.

An Autonomous Drive Balancing Strategy for the Design of Purpose in Open-ended Learning Robots

Distributional reinforcement learning (RL) provides beneficial impacts for the single-agent domain. However, distributional RL methods are not directly compatible with value function factorization methods for multi-agent reinforcement learning. This work provides a distributional perspective on value function factorization, offering a solution for bridging the gap between distributional RL and value function factorization methods.

A Distributional Perspective on Value Function Factorization Methods for Multi-Agent Reinforcement Learning

Several frameworks for decentralized reasoning assume a junction tree agent organization (JT-org). JT-org construction involves 3 related tasks on existence recognition, construction, and environment re-decomposition, where re-decomposition incurs loss of JT-org linked privacy, including privacy on agent, topology, private and shared variables. We propose a novel algorithm DAER that accomplishes all 3 tasks distributively. For Tasks 1 and 2, DAER incurs no loss of JT-org linked privacy.For Task 3, it incurs significantly less privacy loss than existing JT-org construction methods. Its performance is formally analyzed and empirically evaluated.

Constructing Junction Tree Agent Organization with Privacy (JAAMAS Track)

Transportation systems of the future can be best modeled as multi-agent systems. A number of coordination protocols such as autonomous intersection management (AIM), adaptive cooperative traffic light control (TLC), cooperative adaptive cruise control (CACC), among others have been developed with the goal of improving the safety and efficiency of such systems. The overall goal in these systems is to provide behavioral guarantees under the assumption that the participating agents work in concert with a centralized (or distributed) coordinator. While there is work on analyzing such systems from a security perspective, we argue that there is limited work on quantifying trustworthiness of individual agents in a multi-agent system. We propose a framework that uses an epistemic logic to quantify trustworthiness of agents, and embed the use of quantitative trustworthiness values into control and coordination policies. Our modified control policies can help the multi-agent system improve its safety in the presence of untrustworthy agents (and under certain assumptions, including malicious agents). We empirically show the effectiveness of our proposed trust framework by embedding it into AIM, TLC, and CACC platooning algorithms. In our experiments, our trust framework accurately detects attackers in CACC platoons; mitigates the effect of untrustworthy agents in AIM; and trust-aware TLC and AIM reduce collisions in all cases compared to the vanilla versions of these algorithms.

A General Trust Framework for Multi-Agent Systems

In this paper we investigate the effect of an underlying social network over agents in a well-known multi-agent resource allocation problem; the housing market. We first show that, when a housing market takes place over a social network with more than two agents and these agents have an option to avoid forwarding information about it to their followers, there does not exist an exchange mechanism that simultaneously satisfies strategy-proofness, Pareto efficiency, and individual rationality. It is also impossible to find a strategy-proof exchange mechanism that always chooses an outcome in a weakened core. These results highlight the difficulty of taking into account the agents' incentive of information diffusion in the resource allocation. To overcome these negative results, we consider two different ways of restricting the problem; limiting the domain of preferences and the structure of social networks.

Mechanism Design for Housing Markets over Social Networks

Reasoning about observed effects and their causes is important in many applications. For instance, understanding why a plan failed can aid the task of replanning by allowing the agent to tailor a better plan. But under incomplete information, an agent may be unable to determine which actions/events caused an effect. To overcome this, the agent may be able to perform some sensing actions that allow him to figure out what caused the effect. This becomes even more important in multiagent contexts, where an agent may want to identify which agents caused some effect, or possibly prevent other agents from determining who caused something. The effects involved may even be epistemic effects, such as an agent coming to know the PIN of a bank card, and the causes may be sensing actions. Reasoning about such causes is a key part of "theory of mind" and understanding other agents' behaviour. While there has been much work on causality from an objective standpoint, causality from the point of view of individual agents has received much less attention. In this paper, we develop a formalization of knowledge about actual causes in the situation calculus, and how it is affected by actions including sensing. We show that the proposed framework has some intuitive properties and study the conditions under which an agent can be expected to come to know the causes of an effect.

Knowing Why - On the Dynamics of Knowledge about Actual Causes in the Situation Calculus

Several resource allocation problems involve multiple types of resources, with a different agency being responsible for "locally" allocating the resources of each type, while a central planner wishes to provide a guarantee on the properties of the final allocation given agents' preferences. We study the relationship between properties of the local mechanisms, each responsible for assigning all of the resources of a designated type, and the properties of a sequential mechanism which is composed of these local mechanisms, one for each type, applied sequentially, under lexicographic preferences, a well studied model of preferences over multiple types of resources in artificial intelligence and economics. We show that when preferences are O-legal, meaning that agents share a common importance order on the types, sequential mechanisms satisfy the desirable properties of anonymity, neutrality, non-bossiness, or Pareto-optimality if and only if every local mechanism also satisfies the same property, and they are applied sequentially according to the order O. Our main results are that under O-legal lexicographic preferences, every mechanism satisfying strategyproofness and a combination of these properties must be a sequential composition of local mechanisms that are also strategyproof, and satisfy the same combinations of properties.

Multi-Agent Graph-Attention Communication and Teaming

Downloads

Next from AAMAS 2021

Difference Rewards Policy Gradients

Similar lecture

Toward a Self-Learning Governance Loop for Competitive Multi-Attribute MAS

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Multi-Agent Graph-Attention Communication and Teaming

.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 2021

Difference Rewards Policy Gradients

Similar lecture

Toward a Self-Learning Governance Loop for Competitive Multi-Attribute MAS

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads