With the advent of large datasets, offline reinforcement learning (RL) is a promising framework for learning good decision-making policies without the need to interact with the real environment. However, offline RL requires the dataset to be reward-annotated, which presents practical challenges when reward engineering is difficult or when obtaining reward annotations is labor-intensive. In this paper, we introduce Optimal Transport Reward labeling (OTR), an algorithm that assigns rewards to offline trajectories, with a few high-quality demonstrations. OTR's key idea is to use optimal transport to compute an optimal alignment between an unlabeled trajectory in the dataset and an expert demonstration to obtain a similarity measure that can be interpreted as a reward, which can then be used by an offline RL algorithm to learn the policy. OTR is easy to implement and computationally efficient. On D4RL benchmarks, we show that OTR with a single demonstration can consistently match the performance of offline RL with ground-truth rewards.
相關內容
Robotic ultrasound (US) systems have shown great potential to make US examinations easier and more accurate. Recently, various machine learning techniques have been proposed to realize automatic US image interpretation for robotic US acquisition tasks. However, obtaining large amounts of real US imaging data for training is usually expensive or even unfeasible in some clinical applications. An alternative is to build a simulator to generate synthetic US data for training, but the differences between simulated and real US images may result in poor model performance. This work presents a Sim2Real framework to efficiently learn robotic US image analysis tasks based only on simulated data for real-world deployment. A style transfer module is proposed based on unsupervised contrastive learning and used as a preprocessing step to convert the real US images into the simulation style. Thereafter, a task-relevant model is designed to combine CNNs with vision transformers to generate the task-dependent prediction with improved generalization ability. We demonstrate the effectiveness of our method in an image regression task to predict the probe position based on US images in robotic transesophageal echocardiography (TEE). Our results show that using only simulated US data and a small amount of unlabelled real data for training, our method can achieve comparable performance to semi-supervised and fully supervised learning methods. Moreover, the effectiveness of our previously proposed CT-based US image simulation method is also indirectly confirmed.
Contrastive Learning (CL) performances as a rising approach to address the challenge of sparse and noisy recommendation data. Although having achieved promising results, most existing CL methods only perform either hand-crafted data or model augmentation for generating contrastive pairs to find a proper augmentation operation for different datasets, which makes the model hard to generalize. Additionally, since insufficient input data may lead the encoder to learn collapsed embeddings, these CL methods expect a relatively large number of training data (e.g., large batch size or memory bank) to contrast. However, not all contrastive pairs are always informative and discriminative enough for the training processing. Therefore, a more general CL-based recommendation model called Meta-optimized Contrastive Learning for sequential Recommendation (MCLRec) is proposed in this work. By applying both data augmentation and learnable model augmentation operations, this work innovates the standard CL framework by contrasting data and model augmented views for adaptively capturing the informative features hidden in stochastic data augmentation. Moreover, MCLRec utilizes a meta-learning manner to guide the updating of the model augmenters, which helps to improve the quality of contrastive pairs without enlarging the amount of input data. Finally, a contrastive regularization term is considered to encourage the augmentation model to generate more informative augmented views and avoid too similar contrastive pairs within the meta updating. The experimental results on commonly used datasets validate the effectiveness of MCLRec.
The proximal policy optimization (PPO) algorithm stands as one of the most prosperous methods in the field of reinforcement learning (RL). Despite its success, the theoretical understanding of PPO remains deficient. Specifically, it is unclear whether PPO or its optimistic variants can effectively solve linear Markov decision processes (MDPs), which are arguably the simplest models in RL with function approximation. To bridge this gap, we propose an optimistic variant of PPO for episodic adversarial linear MDPs with full-information feedback, and establish a $\tilde{\mathcal{O}}(d^{3/4}H^2K^{3/4})$ regret for it. Here $d$ is the ambient dimension of linear MDPs, $H$ is the length of each episode, and $K$ is the number of episodes. Compared with existing policy-based algorithms, we achieve the state-of-the-art regret bound in both stochastic linear MDPs and adversarial linear MDPs with full information. Additionally, our algorithm design features a novel multi-batched updating mechanism and the theoretical analysis utilizes a new covering number argument of value and policy classes, which might be of independent interest.
Providing a high Quality of Experience (QoE) for video streaming in 5G and beyond 5G (B5G) networks is challenging due to the dynamic nature of the underlying network conditions. Several Adaptive Bit Rate (ABR) algorithms have been developed to improve QoE, but most of them are designed based on fixed rules and unsuitable for a wide range of network conditions. Recently, Deep Reinforcement Learning (DRL) based Asynchronous Advantage Actor-Critic (A3C) methods have recently demonstrated promise in their ability to generalise to diverse network conditions, but they still have limitations. One specific issue with A3C methods is the lag between each actor's behavior policy and central learner's target policy. Consequently, suboptimal updates emerge when the behavior and target policies become out of synchronization. In this paper, we address the problems faced by vanilla-A3C by integrating the on-policy-based multi-agent DRL method into the existing video streaming framework. Specifically, we propose a novel system for ABR generation - Proximal Policy Optimization-based DRL for Adaptive Bit Rate streaming (PPO-ABR). Our proposed method improves the overall video QoE by maximizing sample efficiency using a clipped probability ratio between the new and the old policies on multiple epochs of minibatch updates. The experiments on real network traces demonstrate that PPO-ABR outperforms state-of-the-art methods for different QoE variants.
Imitation is a key component of human social behavior, and is widely used by both children and adults as a way to navigate uncertain or unfamiliar situations. But in an environment populated by multiple heterogeneous agents pursuing different goals or objectives, indiscriminate imitation is unlikely to be an effective strategy -- the imitator must instead determine who is most useful to copy. There are likely many factors that play into these judgements, depending on context and availability of information. Here we investigate the hypothesis that these decisions involve inferences about other agents' reward functions. We suggest that people preferentially imitate the behavior of others they deem to have similar reward functions to their own. We further argue that these inferences can be made on the basis of very sparse or indirect data, by leveraging an inductive bias toward positing the existence of different \textit{groups} or \textit{types} of people with similar reward functions, allowing learners to select imitation targets without direct evidence of alignment.
Reinforcement Learning (RL) algorithms are known to scale poorly to environments with many available actions, requiring numerous samples to learn an optimal policy. The traditional approach of considering the same fixed action space in every possible state implies that the agent must understand, while also learning to maximize its reward, to ignore irrelevant actions such as $\textit{inapplicable actions}$ (i.e. actions that have no effect on the environment when performed in a given state). Knowing this information can help reduce the sample complexity of RL algorithms by masking the inapplicable actions from the policy distribution to only explore actions relevant to finding an optimal policy. While this technique has been formalized for quite some time within the Automated Planning community with the concept of precondition in the STRIPS language, RL algorithms have never formally taken advantage of this information to prune the search space to explore. This is typically done in an ad-hoc manner with hand-crafted domain logic added to the RL algorithm. In this paper, we propose a more systematic approach to introduce this knowledge into the algorithm. We (i) standardize the way knowledge can be manually specified to the agent; and (ii) present a new framework to autonomously learn the partial action model encapsulating the precondition of an action jointly with the policy. We show experimentally that learning inapplicable actions greatly improves the sample efficiency of the algorithm by providing a reliable signal to mask out irrelevant actions. Moreover, we demonstrate that thanks to the transferability of the knowledge acquired, it can be reused in other tasks and domains to make the learning process more efficient.
Deep reinforcement learning algorithms can perform poorly in real-world tasks due to the discrepancy between source and target environments. This discrepancy is commonly viewed as the disturbance in transition dynamics. Many existing algorithms learn robust policies by modeling the disturbance and applying it to source environments during training, which usually requires prior knowledge about the disturbance and control of simulators. However, these algorithms can fail in scenarios where the disturbance from target environments is unknown or is intractable to model in simulators. To tackle this problem, we propose a novel model-free actor-critic algorithm -- namely, state-conservative policy optimization (SCPO) -- to learn robust policies without modeling the disturbance in advance. Specifically, SCPO reduces the disturbance in transition dynamics to that in state space and then approximates it by a simple gradient-based regularizer. The appealing features of SCPO include that it is simple to implement and does not require additional knowledge about the disturbance or specially designed simulators. Experiments in several robot control tasks demonstrate that SCPO learns robust policies against the disturbance in transition dynamics.
Unsupervised domain adaptation has recently emerged as an effective paradigm for generalizing deep neural networks to new target domains. However, there is still enormous potential to be tapped to reach the fully supervised performance. In this paper, we present a novel active learning strategy to assist knowledge transfer in the target domain, dubbed active domain adaptation. We start from an observation that energy-based models exhibit free energy biases when training (source) and test (target) data come from different distributions. Inspired by this inherent mechanism, we empirically reveal that a simple yet efficient energy-based sampling strategy sheds light on selecting the most valuable target samples than existing approaches requiring particular architectures or computation of the distances. Our algorithm, Energy-based Active Domain Adaptation (EADA), queries groups of targe data that incorporate both domain characteristic and instance uncertainty into every selection round. Meanwhile, by aligning the free energy of target data compact around the source domain via a regularization term, domain gap can be implicitly diminished. Through extensive experiments, we show that EADA surpasses state-of-the-art methods on well-known challenging benchmarks with substantial improvements, making it a useful option in the open world. Code is available at //github.com/BIT-DA/EADA.
We propose a new method for event extraction (EE) task based on an imitation learning framework, specifically, inverse reinforcement learning (IRL) via generative adversarial network (GAN). The GAN estimates proper rewards according to the difference between the actions committed by the expert (or ground truth) and the agent among complicated states in the environment. EE task benefits from these dynamic rewards because instances and labels yield to various extents of difficulty and the gains are expected to be diverse -- e.g., an ambiguous but correctly detected trigger or argument should receive high gains -- while the traditional RL models usually neglect such differences and pay equal attention on all instances. Moreover, our experiments also demonstrate that the proposed framework outperforms state-of-the-art methods, without explicit feature engineering.
Recommender systems play a crucial role in mitigating the problem of information overload by suggesting users' personalized items or services. The vast majority of traditional recommender systems consider the recommendation procedure as a static process and make recommendations following a fixed strategy. In this paper, we propose a novel recommender system with the capability of continuously improving its strategies during the interactions with users. We model the sequential interactions between users and a recommender system as a Markov Decision Process (MDP) and leverage Reinforcement Learning (RL) to automatically learn the optimal strategies via recommending trial-and-error items and receiving reinforcements of these items from users' feedbacks. In particular, we introduce an online user-agent interacting environment simulator, which can pre-train and evaluate model parameters offline before applying the model online. Moreover, we validate the importance of list-wise recommendations during the interactions between users and agent, and develop a novel approach to incorporate them into the proposed framework LIRD for list-wide recommendations. The experimental results based on a real-world e-commerce dataset demonstrate the effectiveness of the proposed framework.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191