Multi-agent reinforcement learning (MARL) has achieved notable success in cooperative tasks, demonstrating impressive performance and scalability. However, deploying MARL agents in real-world applications presents critical safety challenges. Current safe MARL algorithms are largely based on the constrained Markov decision process (CMDP) framework, which enforces constraints only on discounted cumulative costs and lacks an all-time safety assurance. Moreover, these methods often overlook the feasibility issue (the system will inevitably violate state constraints within certain regions of the constraint set), resulting in either suboptimal performance or increased constraint violations. To address these challenges, we propose a novel theoretical framework for safe MARL with $\textit{state-wise}$ constraints, where safety requirements are enforced at every state the agents visit. To resolve the feasibility issue, we leverage a control-theoretic notion of the feasible region, the controlled invariant set (CIS), characterized by the safety value function. We develop a multi-agent method for identifying CISs, ensuring convergence to a Nash equilibrium on the safety value function. By incorporating CIS identification into the learning process, we introduce a multi-agent dual policy iteration algorithm that guarantees convergence to a generalized Nash equilibrium in state-wise constrained cooperative Markov games, achieving an optimal balance between feasibility and performance. Furthermore, for practical deployment in complex high-dimensional systems, we propose $\textit{Multi-Agent Dual Actor-Critic}$ (MADAC), a safe MARL algorithm that approximates the proposed iteration scheme within the deep RL paradigm. Empirical evaluations on safe MARL benchmarks demonstrate that MADAC consistently outperforms existing methods, delivering much higher rewards while reducing constraint violations.

With the recent development of natural language generation models - termed as large language models (LLMs) - a potential use case has opened up to improve the way that humans interact with robot assistants. These LLMs should be able to leverage their large breadth of understanding to interpret natural language commands into effective, task appropriate and safe robot task executions. However, in reality, these models suffer from hallucinations, which may cause safety issues or deviations from the task. In other domains, these issues have been improved through the use of collaborative AI systems where multiple LLM agents can work together to collectively plan, code and self-check outputs. In this research, multiple collaborative AI systems were tested against a single independent AI agent to determine whether the success in other domains would translate into improved human-robot interaction performance. The results show that there is no defined trend between the number of agents and the success of the model. However, it is clear that some collaborative AI agent architectures can exhibit a greatly improved capacity to produce error-free code and to solve abstract problems.

Predicting future motions of road participants is an important task for driving autonomously. Most existing models excel at predicting the marginal trajectory of a single agent, but predicting joint trajectories for multiple agents that are consistent within a scene remains a challenge. Previous research has often focused on marginal predictions, but the importance of joint predictions has become increasingly apparent. Joint prediction aims to generate trajectories that are consistent across the entire scene. Our research builds upon the SIMPL baseline to explore methods for generating scene-consistent trajectories. We tested our algorithm on the Argoverse 2 dataset, and experimental results demonstrate that our approach can generate scene-consistent trajectories. Compared to the SIMPL baseline, our method significantly reduces the collision rate of joint trajectories within the scene.

While autonomous agents often surpass humans in their ability to handle vast and complex data, their potential misalignment (i.e., lack of transparency regarding their true objective) has thus far hindered their use in critical applications such as social decision processes. More importantly, existing alignment methods provide no formal guarantees on the safety of such models. Drawing from utility and social choice theory, we provide a novel quantitative definition of alignment in the context of social decision-making. Building on this definition, we introduce probably approximately aligned (i.e., near-optimal) policies, and we derive a sufficient condition for their existence. Lastly, recognizing the practical difficulty of satisfying this condition, we introduce the relaxed concept of safe (i.e., nondestructive) policies, and we propose a simple yet robust method to safeguard the black-box policy of any autonomous agent, ensuring all its actions are verifiably safe for the society.

Matching patients effectively and efficiently for clinical trials is a significant challenge due to the complexity and variability of patient profiles and trial criteria. This paper presents a novel framework, Multi-Agents for Knowledge Augmentation (MAKA), designed to enhance patient-trial matching by dynamically supplementing matching prompts with external, domain-specific knowledge. The MAKA architecture consists of five key components: a knowledge probing agent that detects gaps in domain knowledge, a navigation agent that manages interactions among multiple specialized knowledge augmentation agents, a knowledge augmentation agent that incorporates relevant information into patient-trial matching prompts, a supervision agent aligning the outputs from other agents with the instructions and a matching agent making the final selection decision. This approach enhances the accuracy and contextual richness of patient matching, addresses inherent knowledge gaps in both trail criteria and large language models (LLMs), and improves the alignment between patient characteristics and the criteria.

It has been shown that social institutions impact human motivations to produce different behaviours, such as amount of working or specialisation in labor. With advancement in artificial intelligence (AI), specifically large language models (LLMs), now it is possible to perform in-silico simulations to test various hypotheses around this topic. Here, I simulate two somewhat similar worlds using multi-agent reinforcement learning (MARL) framework of the AI-Economist and generative agent-based model (GABM) framework of the Concordia. In the extended versions of the AI-Economist and Concordia, the agents are able to build houses, trade houses, and trade house building skill. Moreover, along the individualistic-collectivists axis, there are a set of three governing systems: Full-Libertarian, Semi-Libertarian/Utilitarian, and Full-Utilitarian. Additionally, in the extended AI-Economist, the Semi-Libertarian/Utilitarian system is further divided to a set of three governing institutions along the discriminative axis: Inclusive, Arbitrary, and Extractive. Building on these, I am able to show that among governing systems and institutions of the extended AI-Economist, under the Semi-Libertarian/Utilitarian and Inclusive government, the ratios of building houses to trading houses and trading house building skill are higher than the rest. Furthermore, I am able to show that in the extended Concordia when the central government care about equality in the society, the Full-Utilitarian system generates agents building more houses and trading more house building skill. In contrast, these economic activities are higher under the Full-Libertarian system when the central government cares about productivity in the society. Overall, the focus of this paper is to compare and contrast two advanced techniques of AI, MARL and GABM, to simulate a similar social phenomena with limitations.

Research on Reinforcement Learning (RL) approaches for discrete optimization problems has increased considerably, extending RL to an area classically dominated by Operations Research (OR). Vehicle routing problems are a good example of discrete optimization problems with high practical relevance where RL techniques have had considerable success. Despite these advances, open-source development frameworks remain scarce, hampering both the testing of algorithms and the ability to objectively compare results. This ultimately slows down progress in the field and limits the exchange of ideas between the RL and OR communities. Here we propose a library composed of multi-agent environments that simulates classic vehicle routing problems. The library, built on PyTorch, provides a flexible modular architecture design that allows easy customization and incorporation of new routing problems. It follows the Agent Environment Cycle ("AEC") games model and has an intuitive API, enabling rapid adoption and easy integration into existing reinforcement learning frameworks. The library allows for a straightforward use of classical OR benchmark instances in order to narrow the gap between the test beds for algorithm benchmarking used by the RL and OR communities. Additionally, we provide benchmark instance sets for each environment, as well as baseline RL models and training code.

Event-B is a well known methodology for the verified design and development of systems that can be characterised as discrete transition systems. Hybrid Event-B is a conservative extension that interleaves the discrete transitions of Event-B (assumed to be temporally isolated) with episodes of continuously varying state change. While a single Hybrid Event-B machine is sufficient for applications with a single locus of control, it will not do for autonomous systems, which have several loci of control by default. Multi-machine Hybrid Event-B is designed to allow the specification of systems with several loci of control. The formalism is succinctly surveyed, pointing out the subtle semantic issues involved. The multi-machine formalism is then used to specify a relatively simple incident response system, involving a controller, two drones and three responders, working in a partly coordinated and partly independent fashion to manage a putative hazardous scenario.

Behavior Trees (BTs) are high level controllers that have found use in a wide range of robotics tasks. As they grow in popularity and usage, it is crucial to ensure that the appropriate tools and methods are available for ensuring they work as intended. To that end, we created a new methodology by which to create Runtime Monitors for BTs. These monitors can be used by the BT to correct when undesirable behavior is detected and are capable of handling LTL specifications. We demonstrate that in terms of runtime, the generated monitors are on par with monitors generated by existing tools and highlight certain features that make our method more desirable in various situations. We note that our method allows for our monitors to be swapped out with alternate monitors with fairly minimal user effort. Finally, our method ties in with our existing tool, BehaVerify, allowing for the verification of BTs with monitors.

Autonomous systems use independent decision-making with only limited human intervention to accomplish goals in complex and unpredictable environments. As the autonomy technologies that underpin them continue to advance, these systems will find their way into an increasing number of applications in an ever wider range of settings. If we are to deploy them to perform safety-critical or mission-critical roles, it is imperative that we have justified confidence in their safe and correct operation. Verification is the process by which such confidence is established. However, autonomous systems pose challenges to existing verification practices. This paper highlights viewpoints of the Roadmap Working Group of the IEEE Robotics and Automation Society Technical Committee for Verification of Autonomous Systems, identifying these grand challenges, and providing a vision for future research efforts that will be needed to address them.