One of the essential missions in the AI research community is to build an autonomous embodied agent that can attain high-level performance across a wide spectrum of tasks. However, acquiring reward/penalty in all open-ended tasks is unrealistic, making the Reinforcement Learning (RL) training procedure impossible. In this paper, we propose a novel cross-modal contrastive learning framework architecture, CLIP4MC, aiming to learn an RL-friendly vision-language model that serves as a reward function for open-ended tasks. Therefore, no further task-specific reward design is needed. Intuitively, it is more reasonable for the model to address the similarity between the video snippet and the language prompt at both the action and entity levels. To this end, a motion encoder is proposed to capture the motion embeddings across different intervals. The correlation scores are then used to construct the auxiliary reward signal for RL agents. Moreover, we construct a neat YouTube dataset based on the large-scale YouTube database provided by MineDojo. Specifically, two rounds of filtering operations guarantee that the dataset covers enough essential information and that the video-text pair is highly correlated. Empirically, we show that the proposed method achieves better performance on RL tasks compared with baselines.

Building systems that are good for society in the face of complex societal effects requires a dynamic approach. Recent approaches to machine learning (ML) documentation have demonstrated the promise of discursive frameworks for deliberation about these complexities. However, these developments have been grounded in a static ML paradigm, leaving the role of feedback and post-deployment performance unexamined. Meanwhile, recent work in reinforcement learning has shown that the effects of feedback and optimization objectives on system behavior can be wide-ranging and unpredictable. In this paper we sketch a framework for documenting deployed and iteratively updated learning systems, which we call Reward Reports. Taking inspiration from various contributions to the technical literature on reinforcement learning, we outline Reward Reports as living documents that track updates to design choices and assumptions behind what a particular automated system is optimizing for. They are intended to track dynamic phenomena arising from system deployment, rather than merely static properties of models or data. After presenting the elements of a Reward Report, we discuss a concrete example: Meta's BlenderBot 3 chatbot. Several others for game-playing (DeepMind's MuZero), content recommendation (MovieLens), and traffic control (Project Flow) are included in the appendix.

Applying reinforcement learning (RL) to traffic signal control (TSC) has become a promising solution. However, most RL-based methods focus solely on optimization within simulators and give little thought to deployment issues in the real world. Online RL-based methods, which require interaction with the environment, are limited in their interactions with the real-world environment. Additionally, acquiring an offline dataset for offline RL is challenging in the real world. Moreover, most real-world intersections prefer a cyclical phase structure. To address these challenges, we propose: (1) a cyclical offline dataset (COD), designed based on common real-world scenarios to facilitate easy collection; (2) an offline RL model called DataLight, capable of learning satisfactory control strategies from the COD; and (3) a method called Arbitrary To Cyclical (ATC), which can transform most RL-based methods into cyclical signal control. Extensive experiments using real-world datasets on simulators demonstrate that: (1) DataLight outperforms most existing methods and achieves comparable results with the best-performing method; (2) introducing ATC into some recent RL-based methods achieves satisfactory performance; and (3) COD is reliable, with DataLight remaining robust even with a small amount of data. These results suggest that the cyclical offline dataset might be enough for offline RL for TSC. Our proposed methods make significant contributions to the TSC field and successfully bridge the gap between simulation experiments and real-world applications. Our code is released on Github.

Value iteration can find the optimal replenishment policy for a perishable inventory problem, but is computationally demanding due to the large state spaces that are required to represent the age profile of stock. The parallel processing capabilities of modern GPUs can reduce the wall time required to run value iteration by updating many states simultaneously. The adoption of GPU-accelerated approaches has been limited in operational research relative to other fields like machine learning, in which new software frameworks have made GPU programming widely accessible. We used the Python library JAX to implement value iteration and simulators of the underlying Markov decision processes in a high-level API, and relied on this library's function transformations and compiler to efficiently utilize GPU hardware. Our method can extend use of value iteration to settings that were previously considered infeasible or impractical. We demonstrate this on example scenarios from three recent studies which include problems with over 16 million states and additional problem features, such as substitution between products, that increase computational complexity. We compare the performance of the optimal replenishment policies to heuristic policies, fitted using simulation optimization in JAX which allowed the parallel evaluation of multiple candidate policy parameters on thousands of simulated years. The heuristic policies gave a maximum optimality gap of 2.49%. Our general approach may be applicable to a wide range of problems in operational research that would benefit from large-scale parallel computation on consumer-grade GPU hardware.

By enabling agents to communicate, recent cooperative multi-agent reinforcement learning (MARL) methods have demonstrated better task performance and more coordinated behavior. Most existing approaches facilitate inter-agent communication by allowing agents to send messages to each other through free communication channels, i.e., cheap talk channels. Current methods require these channels to be constantly accessible and known to the agents a priori. In this work, we lift these requirements such that the agents must discover the cheap talk channels and learn how to use them. Hence, the problem has two main parts: cheap talk discovery (CTD) and cheap talk utilization (CTU). We introduce a novel conceptual framework for both parts and develop a new algorithm based on mutual information maximization that outperforms existing algorithms in CTD/CTU settings. We also release a novel benchmark suite to stimulate future research in CTD/CTU.

This paper investigates deceptive reinforcement learning for privacy preservation in model-free and continuous action space domains. In reinforcement learning, the reward function defines the agent's objective. In adversarial scenarios, an agent may need to both maximise rewards and keep its reward function private from observers. Recent research presented the ambiguity model (AM), which selects actions that are ambiguous over a set of possible reward functions, via pre-trained $Q$-functions. Despite promising results in model-based domains, our investigation shows that AM is ineffective in model-free domains due to misdirected state space exploration. It is also inefficient to train and inapplicable in continuous action space domains. We propose the deceptive exploration ambiguity model (DEAM), which learns using the deceptive policy during training, leading to targeted exploration of the state space. DEAM is also applicable in continuous action spaces. We evaluate DEAM in discrete and continuous action space path planning environments. DEAM achieves similar performance to an optimal model-based version of AM and outperforms a model-free version of AM in terms of path cost, deceptiveness and training efficiency. These results extend to the continuous domain.

Designing effective routing strategies for mobile wireless networks is challenging due to the need to seamlessly adapt routing behavior to spatially diverse and temporally changing network conditions. In this work, we use deep reinforcement learning (DeepRL) to learn a scalable and generalizable single-copy routing strategy for such networks. We make the following contributions: i) we design a reward function that enables the DeepRL agent to explicitly trade-off competing network goals, such as minimizing delay vs. the number of transmissions per packet; ii) we propose a novel set of relational neighborhood, path, and context features to characterize mobile wireless networks and model device mobility independently of a specific network topology; and iii) we use a flexible training approach that allows us to combine data from all packets and devices into a single offline centralized training set to train a single DeepRL agent. To evaluate generalizeability and scalability, we train our DeepRL agent on one mobile network scenario and then test it on other mobile scenarios, varying the number of devices and transmission ranges. Our results show our learned single-copy routing strategy outperforms all other strategies in terms of delay except for the optimal strategy, even on scenarios on which the DeepRL agent was not trained.

Interactive segmentation has recently been explored to effectively and efficiently harvest high-quality segmentation masks by iteratively incorporating user hints. While iterative in nature, most existing interactive segmentation methods tend to ignore the dynamics of successive interactions and take each interaction independently. We here propose to model iterative interactive image segmentation with a Markov decision process (MDP) and solve it with reinforcement learning (RL) where each voxel is treated as an agent. Considering the large exploration space for voxel-wise prediction and the dependence among neighboring voxels for the segmentation tasks, multi-agent reinforcement learning is adopted, where the voxel-level policy is shared among agents. Considering that boundary voxels are more important for segmentation, we further introduce a boundary-aware reward, which consists of a global reward in the form of relative cross-entropy gain, to update the policy in a constrained direction, and a boundary reward in the form of relative weight, to emphasize the correctness of boundary predictions. To combine the advantages of different types of interactions, i.e., simple and efficient for point-clicking, and stable and robust for scribbles, we propose a supervoxel-clicking based interaction design. Experimental results on four benchmark datasets have shown that the proposed method significantly outperforms the state-of-the-arts, with the advantage of fewer interactions, higher accuracy, and enhanced robustness.

Recently, mean field control (MFC) has provided a tractable and theoretically founded approach to otherwise difficult cooperative multi-agent control. However, the strict assumption of many independent, homogeneous agents may be too stringent in practice. In this work, we propose a novel discrete-time generalization of Markov decision processes and MFC to both many minor agents and potentially complex major agents -- major-minor mean field control (M3FC). In contrast to deterministic MFC, M3FC allows for stochastic minor agent distributions with strong correlation between minor agents through the major agent state, which can model arbitrary problem details not bound to any agent. Theoretically, we give rigorous approximation properties with novel proofs for both M3FC and existing MFC models in the finite multi-agent problem, together with a dynamic programming principle for solving such problems. In the infinite-horizon discounted case, existence of an optimal stationary policy follows. Algorithmically, we propose the major-minor mean field proximal policy optimization algorithm (M3FPPO) as a novel multi-agent reinforcement learning algorithm and demonstrate its success in illustrative M3FC-type problems.

In reinforcement learning, the classic objectives of maximizing discounted and finite-horizon cumulative rewards are PAC-learnable: There are algorithms that learn a near-optimal policy with high probability using a finite amount of samples and computation. In recent years, researchers have introduced objectives and corresponding reinforcement-learning algorithms beyond the classic cumulative rewards, such as objectives specified as linear temporal logic formulas. However, questions about the PAC-learnability of these new objectives have remained open. This work demonstrates the PAC-learnability of general reinforcement-learning objectives through sufficient conditions for PAC-learnability in two analysis settings. In particular, for the analysis that considers only sample complexity, we prove that if an objective given as an oracle is uniformly continuous, then it is PAC-learnable. Further, for the analysis that considers computational complexity, we prove that if an objective is computable, then it is PAC-learnable. In other words, if a procedure computes successive approximations of the objective's value, then the objective is PAC-learnable. We give three applications of our condition on objectives from the literature with previously unknown PAC-learnability and prove that these objectives are PAC-learnable. Overall, our result helps verify existing objectives' PAC-learnability. Also, as some studied objectives that are not uniformly continuous have been shown to be not PAC-learnable, our results could guide the design of new PAC-learnable objectives.