Crowd navigation has received significant research attention in recent years, especially DRL-based methods. While single-robot crowd scenarios have dominated research, they offer limited applicability to real-world complexities. The heterogeneity of interaction among multiple agent categories, like in decentralized multi-robot pedestrian scenarios, are frequently disregarded. This "interaction blind spot" hinders generalizability and restricts progress towards robust navigation algorithms. In this paper, we propose a heterogeneous relational deep reinforcement learning(HeR-DRL), based on customised heterogeneous GNN, in order to improve navigation strategies in decentralized multi-robot crowd navigation. Firstly, we devised a method for constructing robot-crowd heterogenous relation graph that effectively simulates the heterogeneous pair-wise interaction relationships. We proposed a new heterogeneous graph neural network for transferring and aggregating the heterogeneous state information. Finally, we incorporate the encoded information into deep reinforcement learning to explore the optimal policy. HeR-DRL are rigorously evaluated through comparing it to state-of-the-art algorithms in both single-robot and multi-robot circle crowssing scenario. The experimental results demonstrate that HeR-DRL surpasses the state-of-the-art approaches in overall performance, particularly excelling in safety and comfort metrics. This underscores the significance of interaction heterogeneity for crowd navigation. The source code will be publicly released in https://github.com/Zhouxy-Debugging-Den/HeR-DRL.

It is common for us to feel pressure in a competition environment, which arises from the desire to obtain success comparing with other individuals or opponents. Although we might get anxious under the pressure, it could also be a drive for us to stimulate our potentials to the best in order to keep up with others. Inspired by this, we propose a competitive learning framework which is able to help individual robot to acquire knowledge from the competition, fully stimulating its dynamics potential in the race. Specifically, the competition information among competitors is introduced as the additional auxiliary signal to learn advantaged actions. We further build a Multiagent-Race environment, and extensive experiments are conducted, demonstrating that robots trained in competitive environments outperform ones that are trained with SoTA algorithms in single robot environment.

The swift evolution of Large-scale Models (LMs), either language-focused or multi-modal, has garnered extensive attention in both academy and industry. But despite the surge in interest in this rapidly evolving area, there are scarce systematic reviews on their capabilities and potential in distinct impactful scenarios. This paper endeavours to help bridge this gap, offering a thorough examination of the current landscape of LM usage in regards to complex game playing scenarios and the challenges still open. Here, we seek to systematically review the existing architectures of LM-based Agents (LMAs) for games and summarize their commonalities, challenges, and any other insights. Furthermore, we present our perspective on promising future research avenues for the advancement of LMs in games. We hope to assist researchers in gaining a clear understanding of the field and to generate more interest in this highly impactful research direction. A corresponding resource, continuously updated, can be found in our GitHub repository.

Reinforcement Learning- (RL-)based motion planning has recently shown the potential to outperform traditional approaches from autonomous navigation to robot manipulation. In this work, we focus on a motion planning task for an evasive target in a partially observable multi-agent adversarial pursuit-evasion games (PEG). These pursuit-evasion problems are relevant to various applications, such as search and rescue operations and surveillance robots, where robots must effectively plan their actions to gather intelligence or accomplish mission tasks while avoiding detection or capture themselves. We propose a hierarchical architecture that integrates a high-level diffusion model to plan global paths responsive to environment data while a low-level RL algorithm reasons about evasive versus global path-following behavior. Our approach outperforms baselines by 51.2% by leveraging the diffusion model to guide the RL algorithm for more efficient exploration and improves the explanability and predictability.

Compared to other techniques, particle swarm optimization is more frequently utilized because of its ease of use and low variability. However, it is complicated to find the best possible solution in the search space in large-scale optimization problems. Moreover, changing algorithm variables does not influence algorithm convergence much. The PSO algorithm can be combined with other algorithms. It can use their advantages and operators to solve this problem. Therefore, this paper proposes the onlooker multi-parent crossover discrete particle swarm optimization (OMPCDPSO). To improve the efficiency of the DPSO algorithm, we utilized multi-parent crossover on the best solutions. We performed an independent and intensive neighborhood search using the onlooker bees of the bee algorithm. The algorithm uses onlooker bees and crossover. They do local search (exploitation) and global search (exploration). Each of these searches is among the best solutions (employed bees). The proposed algorithm was tested on the allocation problem, which is an NP-hard optimization problem. Also, we used two types of simulated data. They were used to test the scalability and complexity of the better algorithm. Also, fourteen 2D test functions and thirteen 30D test functions were used. They also used twenty IEEE CEC2005 benchmark functions to test the efficiency of OMPCDPSO. Also, to test OMPCDPSO's performance, we compared it to four new binary optimization algorithms and three classic ones. The results show that the OMPCDPSO version had high capability. It performed better than other algorithms. The developed algorithm in this research (OMCDPSO) in 36 test functions out of 47 (76.60%) is better than other algorithms. The Onlooker bees and multi-parent operators significantly impact the algorithm's performance.

The problem of single-agent Evasive AHT (EAHT), Active Hypothesis Testing (AHT) refers to the family of where a passive Eavesdropper (Eve) collects noisy estimates problems where one legitimate agent or decision maker, or a of the legit observations and tries to infer the underlying group of collaborating agents or decision makers, adaptively hypothesis, was studied in [24], focusing however explicitly select(s) sensing actions and collect(s) observations in order on the asymptotical case. In that work, the authors formulated to infer the underlying true hypothesis in a fast and reliable single-agent EAHT as a constrained optimization problem manner [1], [2]. AHT and related problems, such as active including the legitimate agent's and the Eavesdropper's (Eve) parameter estimation [3] and active change point detection [4], error exponent. However, near-optimal or optimal action selection [5], find numerous applications in wireless communications, policies were not presented. In this paper, motivated including anomaly detection over sensor networks [6], strong by the lack of explicit policies for EAHT, we present novel or weak radar models for target detection [7], cyber-intrusion single-and multi-agent EAHT approaches for wireless sensor detection, target search, and adaptive beamforming [8], as well networks that are based on a deep NeuroEvolution (NE) as, very recently, RIS-enabled localization [9] and channel framework. Our contributions are summarized as follows: estimation [10]. In addition, AHT is closely related to the 1) We formulate the single-agent EAHT problem studied feedback channel coding problem [11].

This article explores the dynamic influence of computational entities based on multi-agent systems theory (SMA) combined with large language models (LLM), which are characterized by their ability to simulate complex human interactions, as a possibility to revolutionize human user interaction from the use of specialized artificial agents to support everything from operational organizational processes to strategic decision making based on applied knowledge and human orchestration. Previous investigations reveal that there are limitations, particularly in the autonomous approach of artificial agents, especially when dealing with new challenges and pragmatic tasks such as inducing logical reasoning and problem solving. It is also considered that traditional techniques, such as the stimulation of chains of thoughts, require explicit human guidance. In our approach we employ agents developed from large language models (LLM), each with distinct prototyping that considers behavioral elements, driven by strategies that stimulate the generation of knowledge based on the use case proposed in the scenario (role-play) business, using a discussion approach between agents (guided conversation). We demonstrate the potential of developing agents useful for organizational strategies, based on multi-agent system theories (SMA) and innovative uses based on large language models (LLM based), offering a differentiated and adaptable experiment to different applications, complexities, domains, and capabilities from LLM. Keywords: Multi-Agent Systems (SMA), Artificial Intelligence (AI), Large Language Models (LLM), Artificial Agents

We present a novel class of actor-critic algorithms for actors consisting of sets of interacting modules. We present, analyze theoretically, and empirically evaluate an update rule for each module, which requires only local information: the module's input, output, and the TD error broadcast by a critic. Such updates are necessary when computation of compatible features becomes prohibitively difficult and are also desirable to increase the biological plausibility of reinforcement learning methods.

Multi-Agent Plan Recognition (MAPR) aims to recognize dynamic team structures and team behaviors from the observed team traces (activity sequences) of a set of intelligent agents. Previous MAPR approaches required a library of team activity sequences (team plans) be given as input. However, collecting a library of team plans to ensure adequate coverage is often difficult and costly. In this paper, we relax this constraint, so that team plans are not required to be provided beforehand. We assume instead that a set of action models are available.

In many large economic markets, goods are sold through sequential auctions. Examples include eBay, online ad auctions, wireless spectrum auctions, and the Dutch flower auctions. In this paper, we combine methods from game theory and decision theory to search for approximate equilibria in sequential auction domains, in which bidders do not know their opponents' values for goods, bidders only partially observe the actions of their opponents', and bidders demand multiple goods. We restrict attention to two-phased strategies: first predict (i.e., learn); second, optimize. We use best-reply dynamics [4] for prediction (i.e., to predict other bidders' strategies), and then assuming fixed other-bidder strategies, we estimate and solve the ensuing Markov decision processes (MDP) [18] for optimization. We exploit auction properties to represent the MDP in a more compact state space, and we use Monte Carlo simulation to make estimating the MDP tractable. We show how equilibria found using our search procedure compare to known equilibria for simpler auction domains, and we approximate an equilibrium for a more complex auction domain where analytical solutions are unknown.