Deep reinforcement learning (DRL) provides a promising way for intelligent agents (e.g., autonomous vehicles) to learn to navigate complex scenarios. However, DRL with neural networks as function approximators is typically considered a black box with little explainability and often suffers from suboptimal performance, especially for autonomous navigation in highly interactive multi-agent environments. To address these issues, we propose three auxiliary tasks with spatio-temporal relational reasoning and integrate them into the standard DRL framework, which improves the decision making performance and provides explainable intermediate indicators. We propose to explicitly infer the internal states (i.e., traits and intentions) of surrounding agents (e.g., human drivers) as well as to predict their future trajectories in the situations with and without the ego agent through counterfactual reasoning. These auxiliary tasks provide additional supervision signals to infer the behavior patterns of other interactive agents. Multiple variants of framework integration strategies are compared. We also employ a spatio-temporal graph neural network to encode relations between dynamic entities, which enhances both internal state inference and decision making of the ego agent. Moreover, we propose an interactivity estimation mechanism based on the difference between predicted trajectories in these two situations, which indicates the degree of influence of the ego agent on other agents. To validate the proposed method, we design an intersection driving simulator based on the Intelligent Intersection Driver Model (IIDM) that simulates vehicles and pedestrians. Our approach achieves robust and state-of-the-art performance in terms of standard evaluation metrics and provides explainable intermediate indicators (i.e., internal states, and interactivity scores) for decision making.

This paper examines the effect of real-time, personalized alignment of a robot's reward function to the human's values on trust and team performance. We present and compare three distinct robot interaction strategies: a non-learner strategy where the robot presumes the human's reward function mirrors its own, a non-adaptive-learner strategy in which the robot learns the human's reward function for trust estimation and human behavior modeling, but still optimizes its own reward function, and an adaptive-learner strategy in which the robot learns the human's reward function and adopts it as its own. Two human-subject experiments with a total number of 54 participants were conducted. In both experiments, the human-robot team searches for potential threats in a town. The team sequentially goes through search sites to look for threats. We model the interaction between the human and the robot as a trust-aware Markov Decision Process (trust-aware MDP) and use Bayesian Inverse Reinforcement Learning (IRL) to estimate the reward weights of the human as they interact with the robot. In Experiment 1, we start our learning algorithm with an informed prior of the human's values/goals. In Experiment 2, we start the learning algorithm with an uninformed prior. Results indicate that when starting with a good informed prior, personalized value alignment does not seem to benefit trust or team performance. On the other hand, when an informed prior is unavailable, alignment to the human's values leads to high trust and higher perceived performance while maintaining the same objective team performance.

Deep reinforcement learning excels in various domains but lacks generalizability and interoperability. Programmatic RL methods (Trivedi et al., 2021; Liu et al., 2023) reformulate solving RL tasks as synthesizing interpretable programs that can be executed in the environments. Despite encouraging results, these methods are limited to short-horizon tasks. On the other hand, representing RL policies using state machines (Inala et al., 2020) can inductively generalize to long-horizon tasks; however, it struggles to scale up to acquire diverse and complex behaviors. This work proposes Program Machine Policies (POMPs), which bridge the advantages of programmatic RL and state machine policies, allowing for the representation of complex behaviors and the address of long-term tasks. Specifically, we introduce a method that can retrieve a set of effective, diverse, compatible programs. Then, we use these programs as modes of a state machine and learn a transition function to transition among mode programs, allowing for capturing long-horizon repetitive behaviors. Our proposed framework outperforms programmatic RL and deep RL baselines on various tasks and demonstrates the ability to generalize to even longer horizons without any fine-tuning inductively. Ablation studies justify the effectiveness of our proposed search algorithm for retrieving a set of programs as modes.

Climate change has resulted in a year over year increase in adverse weather and weather conditions which contribute to increasingly severe fire seasons. Without effective mitigation, these fires pose a threat to life, property, ecology, cultural heritage, and critical infrastructure. To better prepare for and react to the increasing threat of wildfires, more accurate fire modelers and mitigation responses are necessary. In this paper, we introduce SimFire, a versatile wildland fire projection simulator designed to generate realistic wildfire scenarios, and SimHarness, a modular agent-based machine learning wrapper capable of automatically generating land management strategies within SimFire to reduce the overall damage to the area. Together, this publicly available system allows researchers and practitioners the ability to emulate and assess the effectiveness of firefighter interventions and formulate strategic plans that prioritize value preservation and resource allocation optimization. The repositories are available for download at https://github.com/mitrefireline.

The optimization of traffic signal control (TSC) is critical for an efficient transportation system. In recent years, reinforcement learning (RL) techniques have emerged as a popular approach for TSC and show promising results for highly adaptive control. However, existing RL-based methods suffer from notably poor real-world applicability and hardly have any successful deployments. The reasons for such failures are mostly due to the reliance on over-idealized traffic simulators for policy optimization, as well as using unrealistic fine-grained state observations and reward signals that are not directly obtainable from real-world sensors. In this paper, we propose a fully Data-Driven and simulator-free framework for realistic Traffic Signal Control (D2TSC). Specifically, we combine well-established traffic flow theory with machine learning to construct a reward inference model to infer the reward signals from coarse-grained traffic data. With the inferred rewards, we further propose a sample-efficient offline RL method to enable direct signal control policy learning from historical offline datasets of real-world intersections. To evaluate our approach, we collect historical traffic data from a real-world intersection, and develop a highly customized simulation environment that strictly follows real data characteristics. We demonstrate through extensive experiments that our approach achieves superior performance over conventional and offline RL baselines, and also enjoys much better real-world applicability.

Wireless communication networks constitute complex systems This interest can be attributed to the innate characteristics of demanding careful optimization of network procedures GNNs, which enable a scalable solution and exhibit inductive to attain predefined performance objectives. Multi-agent deep capability and, thanks to the permutation equivariance property, reinforcement learning (MADRL), owing to its inherent advantages, increased generalization. Notably, these properties find has emerged as a promising strategy for the optimization practical application in works such as [5], where GNNs are of a variety of network problems. Nevertheless, the practical harnessed to capture the dynamic structure of fading channel implementation of MADRL in real systems is hindered by states for the purpose of learning optimal resource allocation challenges related to convergence, which continue to constitute policies in wireless networks. Another domain that has witnessed an active area of research. These challenges encompass the substantial utilization of GNNs is channel management non-stationarity of the environment, the partial observability of within wireless local area networks (WLANs), as evidenced the state, as well as the coordination and cooperation among by works such as [6] and [7]. A notable insight derived from agents [1, 2]. To this end, this paper elucidates the role of the study by Gao et al. [6] is the inherent property of GNNs to leveraging graph structures as an effective means to account provide decentralized inference, rendering them a viable and for non-stationarity in MADRL systems by introducing a promising approach for the practical implementation of overthe-air relational inductive bias in the collective decision-making MADRL systems.

Explainable Reinforcement Learning (XRL) can provide transparency into the decision-making process of a Deep Reinforcement Learning (DRL) model and increase user trust and adoption in real-world use cases. By utilizing XRL techniques, researchers can identify potential vulnerabilities within a trained DRL model prior to deployment, therefore limiting the potential for mission failure or mistakes by the system. This paper introduces the ARLIN (Assured RL Model Interrogation) Toolkit, an open-source Python library that identifies potential vulnerabilities and critical points within trained DRL models through detailed, human-interpretable explainability outputs. To illustrate ARLIN's effectiveness, we provide explainability visualizations and vulnerability analysis for a publicly available DRL model. The open-source code repository is available for download at https://github.com/mitre/arlin.

Large vision-language models are steadily gaining personalization capabilities at the cost of fine-tuning or data augmentation. We present two models for image generation using model-agnostic learning that align semantic priors with generative capabilities. RLDF, or Reinforcement Learning from Diffusion Feedback, is a singular approach for visual imitation through prior-preserving reward function guidance. This employs Q-learning (with standard Q*) for generation and follows a semantic-rewarded trajectory for image search through finite encoding-tailored actions. The second proposed method, noisy diffusion gradient, is optimization driven. At the root of both methods is a special CFG encoding that we propose for continual semantic guidance. Using only a single input image and no text input, RLDF generates high-quality images over varied domains including retail, sports and agriculture showcasing class-consistency and strong visual diversity. Project website is available at https://infernolia.github.io/RLDF.

In this paper, a machine learning-based decentralized time division multiple access (TDMA) algorithm for visible light communication (VLC) Internet of Things (IoT) networks is proposed. The proposed algorithm is based on Q-learning, a reinforcement learning algorithm. This paper considers a decentralized condition in which there is no coordinator node for sending synchronization frames and assigning transmission time slots to other nodes. The proposed algorithm uses a decentralized manner for synchronization, and each node uses the Q-learning algorithm to find the optimal transmission time slot for sending data without collisions. The proposed algorithm is implemented on a VLC hardware system, which had been designed and implemented in our laboratory. Average reward, convergence time, goodput, average delay, and data packet size are evaluated parameters. The results show that the proposed algorithm converges quickly and provides collision-free decentralized TDMA for the network. The proposed algorithm is compared with carrier-sense multiple access with collision avoidance (CSMA/CA) algorithm as a potential selection for decentralized VLC IoT networks. The results show that the proposed algorithm provides up to 61% more goodput and up to 49% less average delay than CSMA/CA.

This report examines whether advanced AIs that perform well in training will be doing so in order to gain power later -- a behavior I call "scheming" (also sometimes called "deceptive alignment"). I conclude that scheming is a disturbingly plausible outcome of using baseline machine learning methods to train goal-directed AIs sophisticated enough to scheme (my subjective probability on such an outcome, given these conditions, is roughly 25%). In particular: if performing well in training is a good strategy for gaining power (as I think it might well be), then a very wide variety of goals would motivate scheming -- and hence, good training performance. This makes it plausible that training might either land on such a goal naturally and then reinforce it, or actively push a model's motivations towards such a goal as an easy way of improving performance. What's more, because schemers pretend to be aligned on tests designed to reveal their motivations, it may be quite difficult to tell whether this has occurred. However, I also think there are reasons for comfort. In particular: scheming may not actually be such a good strategy for gaining power; various selection pressures in training might work against schemer-like goals (for example, relative to non-schemers, schemers need to engage in extra instrumental reasoning, which might harm their training performance); and we may be able to increase such pressures intentionally. The report discusses these and a wide variety of other considerations in detail, and it suggests an array of empirical research directions for probing the topic further.