Learning optimal behavior policy for each agent in multi-agent systems is an essential yet difficult problem. Despite fruitful progress in multi-agent reinforcement learning, the challenge of addressing the dynamics of whether two agents should exhibit consistent behaviors is still under-explored. In this paper, we propose a new approach that enables agents to learn whether their behaviors should be consistent with that of other agents by utilizing intrinsic rewards to learn the optimal policy for each agent. We begin by defining behavior consistency as the divergence in output actions between two agents when provided with the same observation. Subsequently, we introduce dynamic consistency intrinsic reward (DCIR) to stimulate agents to be aware of others' behaviors and determine whether to be consistent with them. Lastly, we devise a dynamic scale network (DSN) that provides learnable scale factors for the agent at every time step to dynamically ascertain whether to award consistent behavior and the magnitude of rewards. We evaluate DCIR in multiple environments including Multi-agent Particle, Google Research Football and StarCraft II Micromanagement, demonstrating its efficacy.

In multi-agent reinforcement learning (MARL), self-interested agents attempt to establish equilibrium and achieve coordination depending on game structure. However, existing MARL approaches are mostly bound by the simultaneous actions of all agents in the Markov game (MG) framework, and few works consider the formation of equilibrium strategies via asynchronous action coordination. In view of the advantages of Stackelberg equilibrium (SE) over Nash equilibrium, we construct a spatio-temporal sequential decision-making structure derived from the MG and propose an N-level policy model based on a conditional hypernetwork shared by all agents. This approach allows for asymmetric training with symmetric execution, with each agent responding optimally conditioned on the decisions made by superior agents. Agents can learn heterogeneous SE policies while still maintaining parameter sharing, which leads to reduced cost for learning and storage and enhanced scalability as the number of agents increases. Experiments demonstrate that our method effectively converges to the SE policies in repeated matrix game scenarios, and performs admirably in immensely complex settings including cooperative tasks and mixed tasks.

This paper proposes a fully scalable multi-agent reinforcement learning (MARL) approach for packet scheduling in conflict graphs, aiming to minimizing average packet delays. Each agent autonomously manages the schedule of a single link over one or multiple sub-bands, considering its own state and states of conflicting links. The problem can be conceptualized as a decentralized partially observable Markov decision process (Dec-POMDP). The proposed solution leverages an on-policy reinforcement learning algorithms multi-agent proximal policy optimization (MAPPO) within a multi-agent networked system, incorporating advanced recurrent structures in the neural network. The MARL design allows for fully decentralized training and execution, seamlessly scaling to very large networks. Extensive simulations across a diverse range of conflict graphs demonstrate that the proposed solution compares favorably to well-established schedulers in terms of both throughput and delay under various traffic conditions.

This paper addresses privacy concerns in multi-agent reinforcement learning (MARL), specifically within the context of supply chains where individual strategic data must remain confidential. Organizations within the supply chain are modeled as agents, each seeking to optimize their own objectives while interacting with others. As each organization's strategy is contingent on neighboring strategies, maintaining privacy of state and action-related information is crucial. To tackle this challenge, we propose a game-theoretic, privacy-preserving mechanism, utilizing a secure multi-party computation (MPC) framework in MARL settings. Our major contribution is the successful implementation of a secure MPC framework, SecFloat on EzPC, to solve this problem. However, simply implementing policy gradient methods such as MADDPG operations using SecFloat, while conceptually feasible, would be programmatically intractable. To overcome this hurdle, we devise a novel approach that breaks down the forward and backward pass of the neural network into elementary operations compatible with SecFloat , creating efficient and secure versions of the MADDPG algorithm. Furthermore, we present a learning mechanism that carries out floating point operations in a privacy-preserving manner, an important feature for successful learning in MARL framework. Experiments reveal that there is on average 68.19% less supply chain wastage in 2 PC compared to no data share, while also giving on average 42.27% better average cumulative revenue for each player. This work paves the way for practical, privacy-preserving MARL, promising significant improvements in secure computation within supply chain contexts and broadly.

Multiwinner voting captures a wide variety of settings, from parliamentary elections in democratic systems to product placement in online shopping platforms. There is a large body of work dealing with axiomatic characterizations, computational complexity, and algorithmic analysis of multiwinner voting rules. Although many challenges remain, significant progress has been made in showing existence of fair and representative outcomes as well as efficient algorithmic solutions for many commonly studied settings. However, much of this work focuses on single-shot elections, even though in numerous real-world settings elections are held periodically and repeatedly. Hence, it is imperative to extend the study of multiwinner voting to temporal settings. Recently, there have been several efforts to address this challenge. However, these works are difficult to compare, as they model multi-period voting in very different ways. We propose a unified framework for studying temporal fairness in this domain, drawing connections with various existing bodies of work, and consolidating them within a general framework. We also identify gaps in existing literature, outline multiple opportunities for future work, and put forward a vision for the future of multiwinner voting in temporal settings.

This paper envisions a revolutionary AIOS-Agent ecosystem, where Large Language Model (LLM) serves as the (Artificial) Intelligent Operating System (IOS, or AIOS)--an operating system "with soul". Upon this foundation, a diverse range of LLM-based AI Agent Applications (Agents, or AAPs) are developed, enriching the AIOS-Agent ecosystem and signaling a paradigm shift from the traditional OS-APP ecosystem. We envision that LLM's impact will not be limited to the AI application level, instead, it will in turn revolutionize the design and implementation of computer system, architecture, software, and programming language, featured by several main concepts: LLM as OS (system-level), Agents as Applications (application-level), Natural Language as Programming Interface (user-level), and Tools as Devices/Libraries (hardware/middleware-level). We begin by introducing the architecture of traditional OS. Then we formalize a conceptual framework for AIOS through "LLM as OS (LLMOS)", drawing analogies between AIOS and traditional OS: LLM is likened to OS kernel, context window to memory, external storage to file system, hardware tools to peripheral devices, software tools to programming libraries, and user prompts to user commands. Subsequently, we introduce the new AIOS-Agent Ecosystem, where users can easily program Agent Applications (AAPs) using natural language, democratizing the development of software, which is different from the traditional OS-APP ecosystem. Following this, we explore the diverse scope of Agent Applications. We delve into both single-agent and multi-agent systems, as well as human-agent interaction. Lastly, drawing on the insights from traditional OS-APP ecosystem, we propose a roadmap for the evolution of the AIOS-Agent ecosystem. This roadmap is designed to guide the future research and development, suggesting systematic progresses of AIOS and its Agent applications.

Artificial intelligence (AI) is being ubiquitously adopted to automate processes in science and industry. However, due to its often intricate and opaque nature, AI has been shown to possess inherent vulnerabilities which can be maliciously exploited with adversarial AI, potentially putting AI users and developers at both cyber and physical risk. In addition, there is insufficient comprehension of the real-world effects of adversarial AI and an inadequacy of AI security examinations; therefore, the growing threat landscape is unknown for many AI solutions. To mitigate this issue, we propose one of the first red team frameworks for evaluating the AI security of maritime autonomous systems. The framework provides operators with a proactive (secure by design) and reactive (post-deployment evaluation) response to securing AI technology today and in the future. This framework is a multi-part checklist, which can be tailored to different systems and requirements. We demonstrate this framework to be highly effective for a red team to use to uncover numerous vulnerabilities within a real-world maritime autonomous systems AI, ranging from poisoning to adversarial patch attacks. The lessons learned from systematic AI red teaming can help prevent MAS-related catastrophic events in a world with increasing uptake and reliance on mission-critical AI.

This study addresses the challenge of online learning in contexts where agents accumulate disparate data, face resource constraints, and use different local algorithms. This paper introduces the Switched Online Learning Algorithm (SOLA), designed to solve the heterogeneous online learning problem by amalgamating updates from diverse agents through a dynamic switching mechanism contingent upon their respective performance and available resources. We theoretically analyze the design of the selecting mechanism to ensure that the regret of SOLA is bounded. Our findings show that the number of changes in selection needs to be bounded by a parameter dependent on the performance of the different local algorithms. Additionally, two test cases are presented to emphasize the effectiveness of SOLA, first on an online linear regression problem and then on an online classification problem with the MNIST dataset.

Formation coordination is a critical aspect of swarm robotics, which involves coordinating the motion and behavior of a group of robots to achieve a specific objective. In formation coordination, the robots must maintain a specific spatial arrangement while in motion. In this paper, we present a leader-follower column formation coordination problem in an unknown, two-dimensional nonlinear manifold, where we redefining it as a trajectory estimation problem. Leveraging Koopman operator theory and Extended Dynamic Mode Decomposition, we estimate the measurement vectors for the follower agent and guide its nonlinear trajectories.

In his "ship of state" allegory (\textit{Republic}, Book VI, 488) Plato poses a question -- how can a crew of sailors presumed to know little about the art of navigation recognize the true pilot among them? The allegory argues that a simple majority voting procedure cannot safely determine who is most qualified to pilot a ship when the voting members are ignorant or biased. We formalize Plato's concerns by considering the problem in AI safety of monitoring noisy AI agents in unsupervised settings. An algorithm evaluating AI agents using unlabeled data would be subject to the evaluation dilemma - how would we know the evaluation algorithm was correct itself? This endless validation chain can be avoided by considering purely algebraic functions of the observed responses. We can construct complete postulates than can prove or disprove the logical consistency of any grading algorithm. A complete set of postulates exists whenever we are evaluating $N$ experts that took $T$ tests with $Q$ questions with $R$ responses each. We discuss evaluating binary classifiers that have taken a single test - the $(N,T=1,Q,R=2)$ tests. We show how some of the postulates have been previously identified in the ML literature but not recognized as such - the \textbf{agreement equations} of Platanios. The complete postulates for pair correlated binary classifiers are considered and we show how it allows for error correlations to be quickly calculated. An algebraic evaluator based on the assumption that the ensemble is error independent is compared with grading by majority voting on evaluations using the \uciadult and and \texttt{two-norm} datasets. Throughout, we demonstrate how the formalism of logical consistency via algebraic postulates of evaluation can help increase the safety of machines using AI algorithms.