Multi-Agent Experience Replay (MER) is a key component of off-policy reinforcement learning~(RL) algorithms. By remembering and reusing experiences from the past, experience replay significantly improves the stability of RL algorithms and their learning efficiency. In many scenarios, multiple agents interact in a shared environment during online training under centralized training and decentralized execution~(CTDE) paradigm. Current multi-agent reinforcement learning~(MARL) algorithms consider experience replay with uniform sampling or based on priority weights to improve transition data sample efficiency in the sampling phase. However, moving transition data histories for each agent through the processor memory hierarchy is a performance limiter. Also, as the agents' transitions continuously renew every iteration, the finite cache capacity results in increased cache misses. To this end, we propose \name, that repeatedly reuses the transitions~(experiences) for a window of $n$ steps in order to improve the cache locality and minimize the transition data movement, instead of sampling new transitions at each step. Specifically, our optimization uses priority weights to select the transitions so that only high-priority transitions will be reused frequently, thereby improving the cache performance. Our experimental results on the Predator-Prey environment demonstrate the effectiveness of reusing the essential transitions based on the priority weights, where we observe an end-to-end training time reduction of $25.4\%$~(for $32$ agents) compared to existing prioritized MER algorithms without notable degradation in the mean reward.

We present the task of "Social Rearrangement", consisting of cooperative everyday tasks like setting up the dinner table, tidying a house or unpacking groceries in a simulated multi-agent environment. In Social Rearrangement, two robots coordinate to complete a long-horizon task, using onboard sensing and egocentric observations, and no privileged information about the environment. We study zero-shot coordination (ZSC) in this task, where an agent collaborates with a new partner, emulating a scenario where a robot collaborates with a new human partner. Prior ZSC approaches struggle to generalize in our complex and visually rich setting, and on further analysis, we find that they fail to generate diverse coordination behaviors at training time. To counter this, we propose Behavior Diversity Play (BDP), a novel ZSC approach that encourages diversity through a discriminability objective. Our results demonstrate that BDP learns adaptive agents that can tackle visual coordination, and zero-shot generalize to new partners in unseen environments, achieving 35% higher success and 32% higher efficiency compared to baselines.

In this study, we explore the emergence of symbols during interactions between individuals through an experimental semiotic study. Previous studies investigate how humans organize symbol systems through communication using artificially designed subjective experiments. In this study, we have focused on a joint attention-naming game (JA-NG) in which participants independently categorize objects and assign names while assuming their joint attention. In the theory of the Metropolis-Hastings naming game (MHNG), listeners accept provided names according to the acceptance probability computed using the Metropolis-Hastings (MH) algorithm. The theory of MHNG suggests that symbols emerge as an approximate decentralized Bayesian inference of signs, which is represented as a shared prior variable if the conditions of MHNG are satisfied. This study examines whether human participants exhibit behavior consistent with MHNG theory when playing JA-NG. By comparing human acceptance decisions of a partner's naming with acceptance probabilities computed in the MHNG, we tested whether human behavior is consistent with the MHNG theory. The main contributions of this study are twofold. First, we reject the null hypothesis that humans make acceptance judgments with a constant probability, regardless of the acceptance probability calculated by the MH algorithm. This result suggests that people followed the acceptance probability computed by the MH algorithm to some extent. Second, the MH-based model predicted human acceptance/rejection behavior more accurately than the other four models: Constant, Numerator, Subtraction, and Binary. This result indicates that symbol emergence in JA-NG can be explained using MHNG and is considered an approximate decentralized Bayesian inference.

In the studies on symbol emergence and emergent communication in a population of agents, a computational model was employed in which agents participate in various language games. Among these, the Metropolis-Hastings naming game (MHNG) possesses a notable mathematical property: symbol emergence through MHNG is proven to be a decentralized Bayesian inference of representations shared by the agents. However, the previously proposed MHNG is limited to a two-agent scenario. This paper extends MHNG to an N-agent scenario. The main contributions of this paper are twofold: (1) we propose the recursive Metropolis-Hastings naming game (RMHNG) as an N-agent version of MHNG and demonstrate that RMHNG is an approximate Bayesian inference method for the posterior distribution over a latent variable shared by agents, similar to MHNG; and (2) we empirically evaluate the performance of RMHNG on synthetic and real image data, enabling multiple agents to develop and share a symbol system. Furthermore, we introduce two types of approximations -- one-sample and limited-length -- to reduce computational complexity while maintaining the ability to explain communication in a population of agents. The experimental findings showcased the efficacy of RMHNG as a decentralized Bayesian inference for approximating the posterior distribution concerning latent variables, which are jointly shared among agents, akin to MHNG. Moreover, the utilization of RMHNG elucidated the agents' capacity to exchange symbols. Furthermore, the study discovered that even the computationally simplified version of RMHNG could enable symbols to emerge among the agents.

Mobile augmented reality (MAR) blends a real scenario with overlaid virtual content, which has been envisioned as one of the ubiquitous interfaces to the Metaverse. Due to the limited computing power and battery life of MAR devices, it is common to offload the computation tasks to edge or cloud servers in close proximity. However, existing offloading solutions developed for MAR tasks suffer from high migration overhead, poor scalability, and short-sightedness when applied in provisioning multi-user MAR services. To address these issues, a MAR service-oriented task offloading scheme is designed and evaluated in edge-cloud computing networks. Specifically, the task interdependency of MAR applications is firstly analyzed and modeled by using directed acyclic graphs. Then, we propose a look-ahead offloading scheme based on a modified Monte Carlo tree (MMCT) search, which can run several multi-step executions in advance to get an estimate of the long-term effect of immediate action. Experiment results show that the proposed offloading scheme can effectively improve the quality of service (QoS) in provisioning multi-user MAR services, compared to four benchmark schemes. Furthermore, it is also shown that the proposed solution is stable and suitable for applications in a highly volatile environment.

Artificial learners often behave differently from human learners in the context of neural agent-based simulations of language emergence and change. A common explanation is the lack of appropriate cognitive biases in these learners. However, it has also been proposed that more naturalistic settings of language learning and use could lead to more human-like results. We investigate this latter account focusing on the word-order/case-marking trade-off, a widely attested language universal that has proven particularly hard to simulate. We propose a new Neural-agent Language Learning and Communication framework (NeLLCom) where pairs of speaking and listening agents first learn a miniature language via supervised learning, and then optimize it for communication via reinforcement learning. Following closely the setup of earlier human experiments, we succeed in replicating the trade-off with the new framework without hard-coding specific biases in the agents. We see this as an essential step towards the investigation of language universals with neural learners.

In the realm of online privacy, privacy assistants play a pivotal role in empowering users to manage their privacy effectively. Although recent studies have shown promising progress in tackling tasks such as privacy violation detection and personalized privacy recommendations, a crucial aspect for widespread user adoption is the capability of these systems to provide explanations for their decision-making processes. This paper presents a privacy assistant for generating explanations for privacy decisions. The privacy assistant focuses on discovering latent topics, identifying explanation categories, establishing explanation schemes, and generating automated explanations. The generated explanations can be used by users to understand the recommendations of the privacy assistant. Our user study of real-world privacy dataset of images shows that users find the generated explanations useful and easy to understand. Additionally, the generated explanations can be used by privacy assistants themselves to improve their decision-making. We show how this can be realized by incorporating the generated explanations into a state-of-the-art privacy assistant.

Large multi-agent systems are often driven by locally defined agent interactions, which is referred to as self-organization. Our primary objective is to determine when the propagation of such local interactions will reach a specific agent of interest. Although conventional approaches that reconstruct all agent states can be used, they may entail unnecessary computational costs. In this paper, we investigate a CNN-LSTM model to forecast the state of a particular agent in a large self-organizing multi-agent system without the reconstruction. The proposed model comprises a CNN encoder to represent the system in a low-dimensional vector, a LSTM module to learn agent dynamics in the vector space, and a MLP decoder to predict the future state of an agent. As an example, we consider a forest fire model where we aim to predict when a particular tree agent will start burning. We compare the proposed model with reconstruction-based approaches such as CNN-LSTM and ConvLSTM. The proposed model exhibits similar or slightly worse AUC but significantly reduces computational costs such as activation than ConvLSTM. Moreover, it achieves higher AUC with less computation than the recontruction-based CNN-LSTM.

The multi-armed bandit (MAB) problem is a paradigm for seque ntial decision-making under uncertainty, which involves allocating a resource to an action, i n order to obtain a reward. MABs address the tradeoff between exploration and exploitation while mak ing sequential decisions. Owing to their utility in large-scale distributed systems (such as inform ation retrieval [ 38 ], advertising [ 8 ], etc.), an extensive study has been conducted on multi-agent versio ns of the classical MAB in the last few years. In multi-agent MABs, there are multiple agents learn ing a bandit and communicating over a network. The goal is to design communication strategies whi ch allow efficient exploration of arms across agents, so that they can perform better than single ag ent MAB algorithms. There exist many versions of multi-agent MABs in the literat ure (see Section 1.2 for an overview). We propose a new collaborative setting where each of the N agents is learning one of M stochastic MABs (where each of the bandits have K arms and M < N) to minimize the group cumulative regret, i.e., the sum of individual cumulative regrets of al l the agents.

Strategic reasoning enables agents to cooperate, communicate, and compete with other agents in diverse situations. Existing approaches to solving strategic games rely on extensive training, yielding strategies that do not generalize to new scenarios or games without retraining. Large Language Models (LLMs), with their ability to comprehend and generate complex, context-rich language, could prove powerful as tools for strategic gameplay. This paper introduces an approach that uses pretrained LLMs with few-shot chain-of-thought examples to enable strategic reasoning for AI agents. Our approach uses systematically generated demonstrations of reasoning about states, values, and beliefs to prompt the model. Using extensive variations of simple matrix games, we show that strategies that are derived based on systematically generated prompts generalize almost perfectly to new game structures, alternate objectives, and hidden information. Additionally, we demonstrate our approach can lead to human-like negotiation strategies in realistic scenarios without any extra training or fine-tuning. Our results highlight the ability of LLMs, guided by systematic reasoning demonstrations, to adapt and excel in diverse strategic scenarios.