Cooperative multi-agent problems often require coordination between agents, which can be achieved through a centralized policy that considers the global state. Multi-agent policy gradient (MAPG) methods are commonly used to learn such policies, but they are often limited to problems with low-level action spaces. In complex problems with large state and action spaces, it is advantageous to extend MAPG methods to use higher-level actions, also known as options, to improve the policy search efficiency. However, multi-robot option executions are often asynchronous, that is, agents may select and complete their options at different time steps. This makes it difficult for MAPG methods to derive a centralized policy and evaluate its gradient, as centralized policy always select new options at the same time. In this work, we propose a novel, conditional reasoning approach to address this problem and demonstrate its effectiveness on representative option-based multi-agent cooperative tasks through empirical validation. Find code and videos at: \href{https://sites.google.com/view/mahrlsupp/}{https://sites.google.com/view/mahrlsupp/}

This white paper presents our work on SurveyLM, a platform for analyzing augmented language models' (ALMs) emergent alignment behaviors through their dynamically evolving attitude and value perspectives in complex social contexts. Social Artificial Intelligence (AI) systems, like ALMs, often function within nuanced social scenarios where there is no singular correct response, or where an answer is heavily dependent on contextual factors, thus necessitating an in-depth understanding of their alignment dynamics. To address this, we apply survey and experimental methodologies, traditionally used in studying social behaviors, to evaluate ALMs systematically, thus providing unprecedented insights into their alignment and emergent behaviors. Moreover, the SurveyLM platform leverages the ALMs' own feedback to enhance survey and experiment designs, exploiting an underutilized aspect of ALMs, which accelerates the development and testing of high-quality survey frameworks while conserving resources. Through SurveyLM, we aim to shed light on factors influencing ALMs' emergent behaviors, facilitate their alignment with human intentions and expectations, and thereby contributed to the responsible development and deployment of advanced social AI systems. This white paper underscores the platform's potential to deliver robust results, highlighting its significance to alignment research and its implications for future social AI systems.

We study a game between liquidity provider and liquidity taker agents interacting in an over-the-counter market, for which the typical example is foreign exchange. We show how a suitable design of parameterized families of reward functions coupled with shared policy learning constitutes an efficient solution to this problem. By playing against each other, our deep-reinforcement-learning-driven agents learn emergent behaviors relative to a wide spectrum of objectives encompassing profit-and-loss, optimal execution and market share. In particular, we find that liquidity providers naturally learn to balance hedging and skewing, where skewing refers to setting their buy and sell prices asymmetrically as a function of their inventory. We further introduce a novel RL-based calibration algorithm which we found performed well at imposing constraints on the game equilibrium. On the theoretical side, we are able to show convergence rates for our multi-agent policy gradient algorithm under a transitivity assumption, closely related to generalized ordinal potential games.

For problems requiring cooperation, many multiagent systems implement solutions among either individual agents or across an entire population towards a common goal. Multiagent teams are primarily studied when in conflict; however, organizational psychology (OP) highlights the benefits of teams among human populations for learning how to coordinate and cooperate. In this paper, we propose a new model of multiagent teams for reinforcement learning (RL) agents inspired by OP and early work on teams in artificial intelligence. We validate our model using complex social dilemmas that are popular in recent multiagent RL and find that agents divided into teams develop cooperative pro-social policies despite incentives to not cooperate. Furthermore, agents are better able to coordinate and learn emergent roles within their teams and achieve higher rewards compared to when the interests of all agents are aligned.

Compartmental epidemiological models categorize individuals based on their disease status, such as the SEIRD model (Susceptible-Exposed-Infected-Recovered-Dead). These models determine the parameters that influence the magnitude of an outbreak, such as contagion and recovery rates. However, they don't account for individual characteristics or population actions, which are crucial for assessing mitigation strategies like mask usage in COVID-19 or condom distribution in HIV. Additionally, studies highlight the role of citizen solidarity, interpersonal trust, and government credibility in explaining differences in contagion rates between countries. Agent-Based Modeling (ABM) offers a valuable approach to study complex systems by simulating individual components, their actions, and interactions within an environment. ABM provides a useful tool for analyzing social phenomena. In this study, we propose an ABM architecture that combines an adapted SEIRD model with a decision-making model for citizens. In this paper, we propose an ABM architecture that allows us to analyze the evolution of virus infections in a society based on two components: 1) an adaptation of the SEIRD model and 2) a decision-making model for citizens. In this way, the evolution of infections is affected, in addition to the spread of the virus itself, by individual behavior when accepting or rejecting public health measures. We illustrate the designed model by examining the progression of SARS-CoV-2 infections in A Coru\~na, Spain. This approach makes it possible to analyze the effect of the individual actions of citizens during an epidemic on the spread of the virus.

Multi-agent systems require effective coordination between groups and individuals to achieve common goals. However, current multi-agent reinforcement learning (MARL) methods primarily focus on improving individual policies and do not adequately address group-level policies, which leads to weak cooperation. To address this issue, we propose a novel Consensus-oriented Strategy (CoS) that emphasizes group and individual policies simultaneously. Specifically, CoS comprises two main components: (a) the vector quantized group consensus module, which extracts discrete latent embeddings that represent the stable and discriminative group consensus, and (b) the group consensus-oriented strategy, which integrates the group policy using a hypernet and the individual policies using the group consensus, thereby promoting coordination at both the group and individual levels. Through empirical experiments on cooperative navigation tasks with both discrete and continuous spaces, as well as Google research football, we demonstrate that CoS outperforms state-of-the-art MARL algorithms and achieves better collaboration, thus providing a promising solution for achieving effective coordination in multi-agent systems.

Connected automated driving has the potential to significantly improve urban traffic efficiency, e.g., by alleviating issues due to occlusion. Cooperative behavior planning can be employed to jointly optimize the motion of multiple vehicles. Most existing approaches to automatic intersection management, however, only consider fully automated traffic. In practice, mixed traffic, i.e., the simultaneous road usage by automated and human-driven vehicles, will be prevalent. The present work proposes to leverage reinforcement learning and a graph-based scene representation for cooperative multi-agent planning. We build upon our previous works that showed the applicability of such machine learning methods to fully automated traffic. The scene representation is extended for mixed traffic and considers uncertainty in the human drivers' intentions. In the simulation-based evaluation, we model measurement uncertainties through noise processes that are tuned using real-world data. The paper evaluates the proposed method against an enhanced first in - first out scheme, our baseline for mixed traffic management. With increasing share of automated vehicles, the learned planner significantly increases the vehicle throughput and reduces the delay due to interaction. Non-automated vehicles benefit virtually alike.

In this work we study a well-known and challenging problem of Multi-agent Pathfinding, when a set of agents is confined to a graph, each agent is assigned a unique start and goal vertices and the task is to find a set of collision-free paths (one for each agent) such that each agent reaches its respective goal. We investigate how to utilize Monte-Carlo Tree Search (MCTS) to solve the problem. Although MCTS was shown to demonstrate superior performance in a wide range of problems like playing antagonistic games (e.g. Go, Chess etc.), discovering faster matrix multiplication algorithms etc., its application to the problem at hand was not well studied before. To this end we introduce an original variant of MCTS, tailored to multi-agent pathfinding. The crux of our approach is how the reward, that guides MCTS, is computed. Specifically, we use individual paths to assist the agents with the the goal-reaching behavior, while leaving them freedom to get off the track if it is needed to avoid collisions. We also use a dedicated decomposition technique to reduce the branching factor of the tree search procedure. Empirically we show that the suggested method outperforms the baseline planning algorithm that invokes heuristic search, e.g.

In order to model criminal networks for law enforcement purposes, a limited supply of data needs to be translated into validated agent-based models. What is missing in current criminological modelling is a systematic and transparent framework for modelers and domain experts that establishes a modelling procedure for computational criminal modelling that includes translating qualitative data into quantitative rules. For this, we propose FREIDA (Framework for Expert-Informed Data-driven Agent-based models). Throughout the paper, the criminal cocaine replacement model (CCRM) will be used as an example case to demonstrate the FREIDA methodology. For the CCRM, a criminal cocaine network in the Netherlands is being modelled where the kingpin node is being removed, the goal being for the remaining agents to reorganize after the disruption and return the network into a stable state. Qualitative data sources such as case files, literature and interviews are translated into empirical laws, and combined with the quantitative sources such as databases form the three dimensions (environment, agents, behaviour) of a networked ABM. Four case files are being modelled and scored both for training as well as for validation scores to transition to the computational model and application phase respectively. In the last phase, iterative sensitivity analysis, uncertainty quantification and scenario testing eventually lead to a robust model that can help law enforcement plan their intervention strategies. Results indicate the need for flexible parameters as well as additional case file simulations to be performed.

Most solutions to the inventory management problem assume a centralization of information that is incompatible with organisational constraints in real supply chain networks. The inventory management problem is a well-known planning problem in operations research, concerned with finding the optimal re-order policy for nodes in a supply chain. While many centralized solutions to the problem exist, they are not applicable to real-world supply chains made up of independent entities. The problem can however be naturally decomposed into sub-problems, each associated with an independent entity, turning it into a multi-agent system. Therefore, a decentralized data-driven solution to inventory management problems using multi-agent reinforcement learning is proposed where each entity is controlled by an agent. Three multi-agent variations of the proximal policy optimization algorithm are investigated through simulations of different supply chain networks and levels of uncertainty. The centralized training decentralized execution framework is deployed, which relies on offline centralization during simulation-based policy identification, but enables decentralization when the policies are deployed online to the real system. Results show that using multi-agent proximal policy optimization with a centralized critic leads to performance very close to that of a centralized data-driven solution and outperforms a distributed model-based solution in most cases while respecting the information constraints of the system.