Hierarchical reinforcement learning is a promising approach that uses temporal abstraction to solve complex long horizon problems. However, simultaneously learning a hierarchy of policies is unstable as it is challenging to train higher-level policy when the lower-level primitive is non-stationary. In this paper, we propose a novel hierarchical algorithm by generating a curriculum of achievable subgoals for evolving lower-level primitives using reinforcement learning and imitation learning. The lower level primitive periodically performs data relabeling on a handful of expert demonstrations using our primitive informed parsing approach. We provide expressions to bound the sub-optimality of our method and develop a practical algorithm for hierarchical reinforcement learning. Since our approach uses a handful of expert demonstrations, it is suitable for most robotic control tasks. Experimental evaluation on complex maze navigation and robotic manipulation environments show that inducing hierarchical curriculum learning significantly improves sample efficiency, and results in efficient goal conditioned policies for solving temporally extended tasks.
相關內容
Offline reinforcement learning (RL) allows agents to learn effective, return-maximizing policies from a static dataset. Three major paradigms for offline RL are Q-Learning, Imitation Learning, and Sequence Modeling. A key open question is: which paradigm is preferred under what conditions? We study this question empirically by exploring the performance of representative algorithms -- Conservative Q-Learning (CQL), Behavior Cloning (BC), and Decision Transformer (DT) -- across the commonly used D4RL and Robomimic benchmarks. We design targeted experiments to understand their behavior concerning data suboptimality and task complexity. Our key findings are: (1) Sequence Modeling requires more data than Q-Learning to learn competitive policies but is more robust; (2) Sequence Modeling is a substantially better choice than both Q-Learning and Imitation Learning in sparse-reward and low-quality data settings; and (3) Sequence Modeling and Imitation Learning are preferable as task horizon increases, or when data is obtained from human demonstrators. Based on the overall strength of Sequence Modeling, we also investigate architectural choices and scaling trends for DT on Atari and D4RL and make design recommendations. We find that scaling the amount of data for DT by 5x gives a 2.5x average score improvement on Atari.
In this work, we first formulate the problem of robotic water scooping using goal-conditioned reinforcement learning. This task is particularly challenging due to the complex dynamics of fluid and the need to achieve multi-modal goals. The policy is required to successfully reach both position goals and water amount goals, which leads to a large convoluted goal state space. To overcome these challenges, we introduce Goal Sampling Adaptation for Scooping (GOATS), a curriculum reinforcement learning method that can learn an effective and generalizable policy for robot scooping tasks. Specifically, we use a goal-factorized reward formulation and interpolate position goal distributions and amount goal distributions to create curriculum throughout the learning process. As a result, our proposed method can outperform the baselines in simulation and achieves 5.46% and 8.71% amount errors on bowl scooping and bucket scooping tasks, respectively, under 1000 variations of initial water states in the tank and a large goal state space. Besides being effective in simulation environments, our method can efficiently adapt to noisy real-robot water-scooping scenarios with diverse physical configurations and unseen settings, demonstrating superior efficacy and generalizability. The videos of this work are available on our project page: //sites.google.com/view/goatscooping.
While unsupervised skill discovery has shown promise in autonomously acquiring behavioral primitives, there is still a large methodological disconnect between task-agnostic skill pretraining and downstream, task-aware finetuning. We present Intrinsic Reward Matching (IRM), which unifies these two phases of learning via the $\textit{skill discriminator}$, a pretraining model component often discarded during finetuning. Conventional approaches finetune pretrained agents directly at the policy level, often relying on expensive environment rollouts to empirically determine the optimal skill. However, often the most concise yet complete description of a task is the reward function itself, and skill learning methods learn an $\textit{intrinsic}$ reward function via the discriminator that corresponds to the skill policy. We propose to leverage the skill discriminator to $\textit{match}$ the intrinsic and downstream task rewards and determine the optimal skill for an unseen task without environment samples, consequently finetuning with greater sample-efficiency. Furthermore, we generalize IRM to sequence skills for complex, long-horizon tasks and demonstrate that IRM enables us to utilize pretrained skills far more effectively than previous skill selection methods on both the Fetch tabletop and Franka Kitchen robot manipulation benchmarks.
Many model-based reinforcement learning (RL) algorithms can be viewed as having two phases that are iteratively implemented: a learning phase where the model is approximately learned and a planning phase where the learned model is used to derive a policy. In the case of standard MDPs, the learning problem can be solved using either value iteration or policy iteration. However, in the case of zero-sum Markov games, there is no efficient policy iteration algorithm; e.g., it has been shown in Hansen et al. (2013) that one has to solve Omega(1/(1-alpha)) MDPs, where alpha is the discount factor, to implement the only known convergent version of policy iteration. Another algorithm for Markov zero-sum games, called naive policy iteration, is easy to implement but is only provably convergent under very restrictive assumptions. Prior attempts to fix naive policy iteration algorithm have several limitations. Here, we show that a simple variant of naive policy iteration for games converges, and converges exponentially fast. The only addition we propose to naive policy iteration is the use of lookahead in the policy improvement phase. This is appealing because lookahead is anyway often used in RL for games. We further show that lookahead can be implemented efficiently in linear Markov games, which are the counterpart of the linear MDPs and have been the subject of much attention recently. We then consider multi-agent reinforcement learning which uses our algorithm in the planning phases, and provide sample and time complexity bounds for such an algorithm.
While distributional reinforcement learning (RL) has demonstrated empirical success, the question of when and why it is beneficial has remained unanswered. In this work, we provide one explanation for the benefits of distributional RL through the lens of small-loss bounds, which scale with the instance-dependent optimal cost. If the optimal cost is small, our bounds are stronger than those from non-distributional approaches. As warmup, we show that learning the cost distribution leads to small-loss regret bounds in contextual bandits (CB), and we find that distributional CB empirically outperforms the state-of-the-art on three challenging tasks. For online RL, we propose a distributional version-space algorithm that constructs confidence sets using maximum likelihood estimation, and we prove that it achieves small-loss regret in the tabular MDPs and enjoys small-loss PAC bounds in latent variable models. Building on similar insights, we propose a distributional offline RL algorithm based on the pessimism principle and prove that it enjoys small-loss PAC bounds, which exhibit a novel robustness property. For both online and offline RL, our results provide the first theoretical benefits of learning distributions even when we only need the mean for making decisions.
We propose a general learning framework for the protection mechanisms that protects privacy via distorting model parameters, which facilitates the trade-off between privacy and utility. The algorithm is applicable to arbitrary privacy measurements that maps from the distortion to a real value. It can achieve personalized utility-privacy trade-off for each model parameter, on each client, at each communication round in federated learning. Such adaptive and fine-grained protection can improve the effectiveness of privacy-preserved federated learning. Theoretically, we show that gap between the utility loss of the protection hyperparameter output by our algorithm and that of the optimal protection hyperparameter is sub-linear in the total number of iterations. The sublinearity of our algorithm indicates that the average gap between the performance of our algorithm and that of the optimal performance goes to zero when the number of iterations goes to infinity. Further, we provide the convergence rate of our proposed algorithm. We conduct empirical results on benchmark datasets to verify that our method achieves better utility than the baseline methods under the same privacy budget.
Auction-based recommender systems are prevalent in online advertising platforms, but they are typically optimized to allocate recommendation slots based on immediate expected return metrics, neglecting the downstream effects of recommendations on user behavior. In this study, we employ reinforcement learning to optimize for long-term return metrics in an auction-based recommender system. Utilizing temporal difference learning, a fundamental reinforcement learning algorithm, we implement an one-step policy improvement approach that biases the system towards recommendations with higher long-term user engagement metrics. This optimizes value over long horizons while maintaining compatibility with the auction framework. Our approach is grounded in dynamic programming ideas which show that our method provably improves upon the existing auction-based base policy. Through an online A/B test conducted on an auction-based recommender system which handles billions of impressions and users daily, we empirically establish that our proposed method outperforms the current production system in terms of long-term user engagement metrics.
Letting robots emulate human behavior has always posed a challenge, particularly in scenarios involving multiple robots. In this paper, we presented a framework aimed at achieving multi-agent reinforcement learning for robot control in construction tasks. The construction industry often necessitates complex interactions and coordination among multiple robots, demanding a solution that enables effective collaboration and efficient task execution. Our proposed framework leverages the principles of proximal policy optimization and developed a multi-agent version to enable the robots to acquire sophisticated control policies. We evaluated the effectiveness of our framework by learning four different collaborative tasks in the construction environments. The results demonstrated the capability of our approach in enabling multiple robots to learn and adapt their behaviors in complex construction tasks while effectively preventing collisions. Results also revealed the potential of combining and exploring the advantages of reinforcement learning algorithms and inverse kinematics. The findings from this research contributed to the advancement of multi-agent reinforcement learning in the domain of construction robotics. By enabling robots to behave like human counterparts and collaborate effectively, we pave the way for more efficient, flexible, and intelligent construction processes.
We introduce DeepNash, an autonomous agent capable of learning to play the imperfect information game Stratego from scratch, up to a human expert level. Stratego is one of the few iconic board games that Artificial Intelligence (AI) has not yet mastered. This popular game has an enormous game tree on the order of $10^{535}$ nodes, i.e., $10^{175}$ times larger than that of Go. It has the additional complexity of requiring decision-making under imperfect information, similar to Texas hold'em poker, which has a significantly smaller game tree (on the order of $10^{164}$ nodes). Decisions in Stratego are made over a large number of discrete actions with no obvious link between action and outcome. Episodes are long, with often hundreds of moves before a player wins, and situations in Stratego can not easily be broken down into manageably-sized sub-problems as in poker. For these reasons, Stratego has been a grand challenge for the field of AI for decades, and existing AI methods barely reach an amateur level of play. DeepNash uses a game-theoretic, model-free deep reinforcement learning method, without search, that learns to master Stratego via self-play. The Regularised Nash Dynamics (R-NaD) algorithm, a key component of DeepNash, converges to an approximate Nash equilibrium, instead of 'cycling' around it, by directly modifying the underlying multi-agent learning dynamics. DeepNash beats existing state-of-the-art AI methods in Stratego and achieved a yearly (2022) and all-time top-3 rank on the Gravon games platform, competing with human expert players.
Training machine learning models in a meaningful order, from the easy samples to the hard ones, using curriculum learning can provide performance improvements over the standard training approach based on random data shuffling, without any additional computational costs. Curriculum learning strategies have been successfully employed in all areas of machine learning, in a wide range of tasks. However, the necessity of finding a way to rank the samples from easy to hard, as well as the right pacing function for introducing more difficult data can limit the usage of the curriculum approaches. In this survey, we show how these limits have been tackled in the literature, and we present different curriculum learning instantiations for various tasks in machine learning. We construct a multi-perspective taxonomy of curriculum learning approaches by hand, considering various classification criteria. We further build a hierarchical tree of curriculum learning methods using an agglomerative clustering algorithm, linking the discovered clusters with our taxonomy. At the end, we provide some interesting directions for future work.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191