This paper focuses on the crucial task of addressing the evacuation of hazardous places, which holds great importance for coordinators, event hosts, and authorities. To facilitate the development of effective solutions, the paper employs Artificial Intelligence (AI) techniques, specifically Multi-Agent Systems (MAS), to construct a simulation model for evacuation. NetLogo is selected as the simulation tool of choice due to its ability to provide a comprehensive understanding of human behaviour in distressing situations within hazardous environments. The primary objective of this paper is to enhance our comprehension of how individuals react and respond during such distressing situations. By leveraging AI and MAS, the simulation model aims to capture the complex dynamics of evacuation scenarios, enabling policymakers and emergency planners to make informed decisions and implement more efficient and effective evacuation strategies. This paper endeavours to contribute to the advancement of evacuation planning and ultimately improve the safety and well-being of individuals in hazardous places

While speech-enabled teachable agents have some advantages over typing-based ones, they are vulnerable to errors stemming from misrecognition by automatic speech recognition (ASR). These errors may propagate, resulting in unexpected changes in the flow of conversation. We analyzed how such changes are linked with learning gains and learners' rapport with the agents. Our results show they are not related to learning gains or rapport, regardless of the types of responses the agents should have returned given the correct input from learners without ASR errors. We also discuss the implications for optimal error-recovery policies for teachable agents that can be drawn from these findings.

Federated learning (FL) is a recent machine learning framework that allows multiple agents, each of them with their own dataset, to train a model collaboratively without sharing their data [1-4]. The federated setting assumes that all agents are connected to a server that can communicate with each of them and that is in charge of aggregating the agents' updates to obtain the global model. This is similar to parallel distributed (PD) model training [5-8], with one crucial difference: in the latter, the agents send gradients to the central server to update the parameter value with a gradient step, while in FL the agents send their own local parameters for the server to average them. This has an impact on the communication frequency required by each framework: in PD one round of communication between (usually all) the agents and the server has to happen every time a (mini-batch) stochastic gradient descent (SGD) step is taken at the nodes, while in FL (i) multiple SGD updates can happen before a new server communication round takes place (which in FL literature are usually called local updates), and (ii) not all devices need to engage in the server communication round (which is known as partial participation). This makes FL a much more suitable option for settings with a large number of agents and a limited communication bandwidth with the server. In contrast to the approaches described above, the decentralized setting does not rely on a central server for the aggregation of the nodes' updates.

Planning is a data efficient decision-making strategy where an agent selects candidate actions by exploring possible future states. To simulate future states when there is a high-dimensional action space, the knowledge of one's decision making strategy must be used to limit the number of actions to be explored. We refer to the model used to simulate one's decisions as the agent's self-model. While self-models are implicitly used widely in conjunction with world models to plan actions, it remains unclear how self-models should be designed. Inspired by current reinforcement learning approaches and neuroscience, we explore the benefits and limitations of using a distilled policy network as the self-model. In such dual-policy agents, a model-free policy and a distilled policy are used for model-free actions and planned actions, respectively. Our results on a ecologically relevant, parametric environment indicate that distilled policy network for self-model stabilizes training, has faster inference than using model-free policy, promotes better exploration, and could learn a comprehensive understanding of its own behaviors, at the cost of distilling a new network apart from the model-free policy.

This study probes the vulnerabilities of cooperative multi-agent reinforcement learning (c-MARL) under adversarial attacks, a critical determinant of c-MARL's worst-case performance prior to real-world implementation. Current observation-based attacks, constrained by white-box assumptions, overlook c-MARL's complex multi-agent interactions and cooperative objectives, resulting in impractical and limited attack capabilities. To address these shortcomes, we propose Adversarial Minority Influence (AMI), a practical and strong for c-MARL. AMI is a practical black-box attack and can be launched without knowing victim parameters. AMI is also strong by considering the complex multi-agent interaction and the cooperative goal of agents, enabling a single adversarial agent to unilaterally misleads majority victims to form targeted worst-case cooperation. This mirrors minority influence phenomena in social psychology. To achieve maximum deviation in victim policies under complex agent-wise interactions, our unilateral attack aims to characterize and maximize the impact of the adversary on the victims. This is achieved by adapting a unilateral agent-wise relation metric derived from mutual information, thereby mitigating the adverse effects of victim influence on the adversary. To lead the victims into a jointly detrimental scenario, our targeted attack deceives victims into a long-term, cooperatively harmful situation by guiding each victim towards a specific target, determined through a trial-and-error process executed by a reinforcement learning agent. Through AMI, we achieve the first successful attack against real-world robot swarms and effectively fool agents in simulated environments into collectively worst-case scenarios, including Starcraft II and Multi-agent Mujoco. The source code and demonstrations can be found at: https://github.com/DIG-Beihang/AMI.

We study the parameterized complexity of winner determination problems for three prevalent $k$-committee selection rules, namely the minimax approval voting (MAV), the proportional approval voting (PAV), and the Chamberlin-Courant's approval voting (CCAV). It is known that these problems are computationally hard. Although they have been studied from the parameterized complexity point of view with respect to several natural parameters, many of them turned out to be W[1]-hard or W[2]-hard. Aiming at obtaining plentiful fixed-parameter algorithms, we revisit these problems by considering more natural single parameters, combined parameters, and structural parameters.

The widely adopted spatial modeling method -- standard Gaussian process (GP) regression -- becomes inadequate in processing the growing sensing data of a large size. To overcome the computational challenge, this paper presents an environmental modeling framework using a sparse variant of GP called streaming sparse GP (SSGP). The SSGP is able to handle streaming data in an online and incremental manner, and is therefore suitable for long-term autonomous environmental monitoring. The SSGP summarizes the collected data using a small set of pseudo data points that best represent the whole dataset, and updates the hyperparameters and pseudo point locations in a streaming fashion, leading to high-quality approximation of the underlying environmental model with significantly reduced computational cost and memory demand.

This thesis studies the domain of collective robotics, and more particularly the optimization problems of multirobot systems in the context of exploration, path planning and coordination. It includes two contributions. The first one is the use of the Butterfly Optimization Algorithm (BOA) to solve the Unknown Area Exploration problem with energy constraints in dynamic environments. This algorithm was never used for solving robotics problems before, as far as we know. We proposed a new version of this algorithm called xBOA based on the crossover operator to improve the diversity of the candidate solutions and speed up the convergence of the algorithm. The second contribution is the development of a new simulation framework for benchmarking dynamic incremental problems in robotics such as exploration tasks. The framework is made in such a manner to be generic to quickly compare different metaheuristics with minimum modifications, and to adapt easily to single and multi-robot scenarios. Also, it provides researchers with tools to automate their experiments and generate visuals, which will allow them to focus on more important tasks such as modeling new algorithms. We conducted a series of experiments that showed promising results and allowed us to validate our approach and model.

The Flatland Challenge, which was first held in 2019 and reported in NeurIPS 2020, is designed to answer the question: How to efficiently manage dense traffic on complex rail networks? Considering the significance of punctuality in real-world railway network operation and the fact that fast passenger trains share the network with slow freight trains, Flatland version 3 introduces trains with different speeds and scheduling time windows. This paper introduces the Flatland 3 problem definitions and extends an award-winning MAPF-based software, which won the NeurIPS 2020 competition, to efficiently solve Flatland 3 problems. The resulting system won the Flatland 3 competition. We designed a new priority ordering for initial planning, a new neighbourhood selection strategy for efficient solution quality improvement with Multi-Agent Path Finding via Large Neighborhood Search(MAPF-LNS), and use MAPF-LNS for partially replanning the trains influenced by malfunction.

This work deals with the heterogeneous semantic-native communication (SNC) problem. When agents do not share the same communication context, the effectiveness of contextual reasoning (CR) is compromised calling for agents to infer other agents' context. This article proposes a novel framework for solving the inverse problem of CR in SNC using two Bayesian inference methods, namely: Bayesian inverse CR (iCR) and Bayesian inverse linearized CR (iLCR). The first proposed Bayesian iCR method utilizes Markov Chain Monte Carlo (MCMC) sampling to infer the agent's context while being computationally expensive. To address this issue, a Bayesian iLCR method is leveraged which obtains a linearized CR (LCR) model by training a linear neural network. Experimental results show that the Bayesian iLCR method requires less computation and achieves higher inference accuracy compared to Bayesian iCR. Additionally, heterogeneous SNC based on the context obtained through the Bayesian iLCR method shows better communication effectiveness than that of Bayesian iCR. Overall, this work provides valuable insights and methods to improve the effectiveness of SNC in situations where agents have different contexts.