With the constraint of a no regret follower, will the players in a two-player Stackelberg game still reach Stackelberg equilibrium? We first show when the follower strategy is either reward-average or transform-reward-average, the two players can always get the Stackelberg Equilibrium. Then, we extend that the players can achieve the Stackelberg equilibrium in the two-player game under the no regret constraint. Also, we show a strict upper bound of the follower's utility difference between with and without no regret constraint. Moreover, in constant-sum two-player Stackelberg games with non-regret action sequences, we ensure the total optimal utility of the game remains also bounded.

Efficient navigation in dynamic environments is crucial for autonomous robots interacting with various environmental entities, including both moving agents and static obstacles. In this study, we present a novel methodology that enhances the robot's interaction with different types of agents and obstacles based on specific safety requirements. This approach uses information about the entity types, improving collision avoidance and ensuring safer navigation. We introduce a new reward function that penalizes the robot for collisions with different entities such as adults, bicyclists, children, and static obstacles, and additionally encourages the robot's proximity to the goal. It also penalizes the robot for being close to entities, and the safe distance also depends on the entity type. Additionally, we propose an optimized algorithm for training and testing, which significantly accelerates train, validation, and test steps and enables training in complex environments. Comprehensive experiments conducted using simulation demonstrate that our approach consistently outperforms conventional navigation and collision avoidance methods, including state-of-the-art techniques. To sum up, this work contributes to enhancing the safety and efficiency of navigation systems for autonomous robots in dynamic, crowded environments.

Reinforcement Learning (RL) training is predominantly conducted in cost-effective and controlled simulation environments. However, the transfer of these trained models to real-world tasks often presents unavoidable challenges. This research explores the direct training of RL algorithms in controlled yet realistic real-world settings for the execution of dexterous manipulation. The benchmarking results of three RL algorithms trained on intricate in-hand manipulation tasks within practical real-world contexts are presented. Our study not only demonstrates the practicality of RL training in authentic real-world scenarios, facilitating direct real-world applications, but also provides insights into the associated challenges and considerations. Additionally, our experiences with the employed experimental methods are shared, with the aim of empowering and engaging fellow researchers and practitioners in this dynamic field of robotics.

In actor-critic-based reinforcement learning algorithms such as Twin Delayed Deep Deterministic policy gradient (TD3), insufficient exploration of the spatial space can result in suboptimal policies when controlling 7-DOF robotic arms. To address this issue, we propose a novel Exploration-Enhanced Contrastive Learning (EECL) module that improves exploration by providing additional rewards for encountering novel states. Our module stores previously explored states in a buffer and identifies new states by comparing them with historical data using Euclidean distance within a K-dimensional tree (KDTree) framework. When the agent explores new states, exploration rewards are assigned. These rewards are then integrated into the TD3 algorithm, ensuring that the Q-learning process incorporates these signals, promoting more effective strategy optimization. We evaluate our method on the robosuite panda lift task, demonstrating that it significantly outperforms the baseline TD3 in terms of both efficiency and convergence speed in the tested environment.

Centralized Training for Decentralized Execution, where agents are trained offline in a centralized fashion and execute online in a decentralized manner, has become a popular approach in Multi-Agent Reinforcement Learning (MARL). In particular, it has become popular to develop actor-critic methods that train decentralized actors with a centralized critic where the centralized critic is allowed access global information of the entire system, including the true system state. Such centralized critics are possible given offline information and are not used for online execution. While these methods perform well in a number of domains and have become a de facto standard in MARL, using a centralized critic in this context has yet to be sufficiently analyzed theoretically or empirically. In this paper, we therefore formally analyze centralized and decentralized critic approaches, and analyze the effect of using state-based critics in partially observable environments. We derive theories contrary to the common intuition: critic centralization is not strictly beneficial, and using state values can be harmful. We further prove that, in particular, state-based critics can introduce unexpected bias and variance compared to history-based critics. Finally, we demonstrate how the theory applies in practice by comparing different forms of critics on a wide range of common multi-agent benchmarks. The experiments show practical issues such as the difficulty of representation learning with partial observability, which highlights why the theoretical problems are often overlooked in the literature.

Data efficiency of learning, which plays a key role in the Reinforcement Learning (RL) training process, becomes even more important in continual RL with sequential environments. In continual RL, the learner interacts with non-stationary, sequential tasks and is required to learn new tasks without forgetting previous knowledge. However, there is little work on implementing data augmentation for continual RL. In this paper, we investigate the efficacy of data augmentation for continual RL. Specifically, we provide benchmarking data augmentations for continual RL, by (1) summarising existing data augmentation methods and (2) including a new augmentation method for continual RL: Adversarial Augmentation with Gradient Episodic Memory (Adv-GEM). Extensive experiments show that data augmentations, such as random amplitude scaling, state-switch, mixup, adversarial augmentation, and Adv-GEM, can improve existing continual RL algorithms in terms of their average performance, catastrophic forgetting, and forward transfer, on robot control tasks. All data augmentation methods are implemented as plug-in modules for trivial integration into continual RL methods.

Recently, reinforcement learning (RL) has proved a promising alternative for conventional local heuristics in score-based approaches to learning directed acyclic causal graphs (DAGs) from observational data. However, the intricate acyclicity constraint still challenges the efficient exploration of the vast space of DAGs in existing methods. In this study, we introduce ALIAS (reinforced dAg Learning wIthout Acyclicity conStraints), a novel approach to causal discovery powered by the RL machinery. Our method features an efficient policy for generating DAGs in just a single step with an optimal quadratic complexity, fueled by a novel parametrization of DAGs that directly translates a continuous space to the space of all DAGs, bypassing the need for explicitly enforcing acyclicity constraints. This approach enables us to navigate the search space more effectively by utilizing policy gradient methods and established scoring functions. In addition, we provide compelling empirical evidence for the strong performance of ALIAS in comparison with state-of-the-arts in causal discovery over increasingly difficult experiment conditions on both synthetic and real datasets.

Reinforcement learning is a widely used approach to autonomous navigation, showing potential in various tasks and robotic setups. Still, it often struggles to reach distant goals when safety constraints are imposed (e.g., the wheeled robot is prohibited from moving close to the obstacles). One of the main reasons for poor performance in such setups, which is common in practice, is that the need to respect the safety constraints degrades the exploration capabilities of an RL agent. To this end, we introduce a novel learnable algorithm that is based on decomposing the initial problem into smaller sub-problems via intermediate goals, on the one hand, and respects the limit of the cumulative safety constraints, on the other hand -- SPEIS(Safe Policy Exploration Improvement via Subgoals). It comprises the two coupled policies trained end-to-end: subgoal and safe. The subgoal policy is trained to generate the subgoal based on the transitions from the buffer of the safe (main) policy that helps the safe policy to reach distant goals. Simultaneously, the safe policy maximizes its rewards while attempting not to violate the limit of the cumulative safety constraints, thus providing a certain level of safety. We evaluate SPEIS in a wide range of challenging (simulated) environments that involve different types of robots in two different environments: autonomous vehicles from the POLAMP environment and car, point, doggo, and sweep from the safety-gym environment. We demonstrate that our method consistently outperforms state-of-the-art competitors and can significantly reduce the collision rate while maintaining high success rates (higher by 80% compared to the best-performing methods).

Decentralized stochastic control problems are intrinsically difficult to study because of the inapplicability of standard tools from centralized control such as dynamic programming and the resulting computational complexity. In this paper, we address some of these challenges for decentralized stochastic control with Borel spaces under three different but tightly related information structures under a unified theme: the one-step delayed information sharing pattern, the K-step periodic information sharing pattern, and the completely decentralized information structure where no sharing of information occurs. We will show that the one-step delayed and K-step periodic problems can be reduced to a centralized MDP, generalizing prior results which considered finite, linear, or static models, by addressing several measurability questions. The separated nature of policies under both information structures is then established. We then provide sufficient conditions for the transition kernels of both centralized reductions to be weak-Feller, which facilitates rigorous approximation and learning theoretic results. We will then show that for the completely decentralized control problem finite memory local policies are near optimal under a joint conditional mixing condition. This is achieved by obtaining a bound for finite memory policies which goes to zero as memory size increases. We will also provide a performance bound for the K-periodic problem, which results from replacing the full common information by a finite sliding window of information. The latter will depend on the condition of predictor stability in expected total variation, which we will establish. We finally show that under the periodic information sharing pattern, a quantized Q-learning algorithm converges asymptotically towards a near optimal solution. Each of the above, to our knowledge, is a new contribution to the literature.

The emergence of mobile robotics, particularly in the automotive industry, introduces a promising era of enriched user experiences and adept handling of complex navigation challenges. The realization of these advancements necessitates a focused technological effort and the successful execution of numerous intricate tasks, particularly in the critical domain of Simultaneous Localization and Mapping (SLAM). Various artificial intelligence (AI) methodologies, such as deep learning and reinforcement learning, present viable solutions to address the challenges in SLAM. This study specifically explores the application of reinforcement learning in the context of SLAM. By enabling the agent (the robot) to iteratively interact with and receive feedback from its environment, reinforcement learning facilitates the acquisition of navigation and mapping skills, thereby enhancing the robot's decision-making capabilities. This approach offers several advantages, including improved navigation proficiency, increased resilience, reduced dependence on sensor precision, and refinement of the decision-making process. The findings of this study, which provide an overview of reinforcement learning's utilization in SLAM, reveal significant advancements in the field. The investigation also highlights the evolution and innovative integration of these techniques.