PyTorchKR Papers

최신 논문

총 13,741편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2310.10076

Verbosity Bias in Preference Labeling by Large Language Models

In recent years, Large Language Models (LLMs) have witnessed a remarkable surge in prevalence, altering the landscape of natural language processing and machine learning. One key factor in improving the performance of LLMs is alignment with humans achieved with Reinforcement Learning from Human Feedback (RLHF), as for many LLMs such as GPT-4, Bard, etc. In addition, recent studies are investigating the replacement of human feedback with feedback from other LLMs named Reinforcement Learning from AI Feedback (RLAIF). We examine the biases that come along with evaluating LLMs with other LLMs and take a closer look into verbosity bias -- a bias where LLMs sometimes prefer more verbose answers even if they have similar qualities. We see that in our problem setting, GPT-4 prefers longer answers more than humans. We also propose a metric to measure this bias.

arXiv:2310.09858

Federated Reinforcement Learning for Resource Allocation in V2X Networks

Resource allocation significantly impacts the performance of vehicle-to-everything (V2X) networks. Most existing algorithms for resource allocation are based on optimization or machine learning (e.g., reinforcement learning). In this paper, we explore resource allocation in a V2X network under the framework of federated reinforcement learning (FRL). On one hand, the usage of RL overcomes many challenges from the model-based optimization schemes. On the other hand, federated learning (FL) enables agents to deal with a number of practical issues, such as privacy, communication overhead, and exploration efficiency. The framework of FRL is then implemented by the inexact alternative direction method of multipliers (ADMM), where subproblems are solved approximately using policy gradients and accelerated by an adaptive step size calculated from their second moments. The developed algorithm, PASM, is proven to be convergent under mild conditions and has a nice numerical performance compared with some baseline methods for solving the resource allocation problem in a V2X network.

arXiv:2310.09833

Robust Multi-Agent Reinforcement Learning by Mutual Information Regularization

In multi-agent reinforcement learning (MARL), ensuring robustness against unpredictable or worst-case actions by allies is crucial for real-world deployment. Existing robust MARL methods either approximate or enumerate all possible threat scenarios against worst-case adversaries, leading to computational intensity and reduced robustness. In contrast, human learning efficiently acquires robust behaviors in daily life without preparing for every possible threat. Inspired by this, we frame robust MARL as an inference problem, with worst-case robustness implicitly optimized under all threat scenarios via off-policy evaluation. Within this framework, we demonstrate that Mutual Information Regularization as Robust Regularization (MIR3) during routine training is guaranteed to maximize a lower bound on robustness, without the need for adversaries. Further insights show that MIR3 acts as an information bottleneck, preventing agents from over-reacting to others and aligning policies with robust action priors. In the presence of worst-case adversaries, our MIR3 significantly surpasses baseline methods in robustness and training efficiency while maintaining cooperative performance in StarCraft II and robot swarm control. When deploying the robot swarm control algorithm in the real world, our method also outperforms the best baseline by 14.29%.

arXiv:2310.10312

End-to-end Offline Reinforcement Learning for Glycemia Control

The development of closed-loop systems for glycemia control in type I diabetes relies heavily on simulated patients. Improving the performances and adaptability of these close-loops raises the risk of over-fitting the simulator. This may have dire consequences, especially in unusual cases which were not faithfully-if at all-captured by the simulator. To address this, we propose to use offline RL agents, trained on real patient data, to perform the glycemia control. To further improve the performances, we propose an end-to-end personalization pipeline, which leverages offline-policy evaluation methods to remove altogether the need of a simulator, while still enabling an estimation of clinically relevant metrics for diabetes.

arXiv:2310.08602공식 코드 · Papers with Code

Safe Deep Policy Adaptation

A critical goal of autonomy and artificial intelligence is enabling autonomous robots to rapidly adapt in dynamic and uncertain environments. Classic adaptive control and safe control provide stability and safety guarantees but are limited to specific system classes. In contrast, policy adaptation based on reinforcement learning (RL) offers versatility and generalizability but presents safety and robustness challenges. We propose SafeDPA, a novel RL and control framework that simultaneously tackles the problems of policy adaptation and safe reinforcement learning. SafeDPA jointly learns adaptive policy and dynamics models in simulation, predicts environment configurations, and fine-tunes dynamics models with few-shot real-world data. A safety filter based on the Control Barrier Function (CBF) on top of the RL policy is introduced to ensure safety during real-world deployment. We provide theoretical safety guarantees of SafeDPA and show the robustness of SafeDPA against learning errors and extra perturbations. Comprehensive experiments on (1) classic control problems (Inverted Pendulum), (2) simulation benchmarks (Safety Gym), and (3) a real-world agile robotics platform (RC Car) demonstrate great superiority of SafeDPA in both safety and task performance, over state-of-the-art baselines. Particularly, SafeDPA demonstrates notable generalizability, achieving a 300% increase in safety rate compared to the baselines, under unseen disturbances in real-world experiments.

arXiv:2310.08595

Deep Reinforcement Learning for Autonomous Vehicle Intersection Navigation

In this paper, we explore the challenges associated with navigating complex T-intersections in dense traffic scenarios for autonomous vehicles (AVs). Reinforcement learning algorithms have emerged as a promising approach to address these challenges by enabling AVs to make safe and efficient decisions in real-time. Here, we address the problem of efficiently and safely navigating T-intersections using a lower-cost, single-agent approach based on the Twin Delayed Deep Deterministic Policy Gradient (TD3) reinforcement learning algorithm. We show that our TD3-based method, when trained and tested in the CARLA simulation platform, demonstrates stable convergence and improved safety performance in various traffic densities. Our results reveal that the proposed approach enables the AV to effectively navigate T-intersections, outperforming previous methods in terms of travel delays, collision minimization, and overall cost. This study contributes to the growing body of knowledge on reinforcement learning applications in autonomous driving and highlights the potential of single-agent, cost-effective methods for addressing more complex driving scenarios and advancing reinforcement learning algorithms in the future.

arXiv:2310.09144

Goodhart's Law in Reinforcement Learning

Implementing a reward function that perfectly captures a complex task in the real world is impractical. As a result, it is often appropriate to think of the reward function as a proxy for the true objective rather than as its definition. We study this phenomenon through the lens of Goodhart's law, which predicts that increasing optimisation of an imperfect proxy beyond some critical point decreases performance on the true objective. First, we propose a way to quantify the magnitude of this effect and show empirically that optimising an imperfect proxy reward often leads to the behaviour predicted by Goodhart's law for a wide range of environments and reward functions. We then provide a geometric explanation for why Goodhart's law occurs in Markov decision processes. We use these theoretical insights to propose an optimal early stopping method that provably avoids the aforementioned pitfall and derive theoretical regret bounds for this method. Moreover, we derive a training method that maximises worst-case reward, for the setting where there is uncertainty about the true reward function. Finally, we evaluate our early stopping method experimentally. Our results support a foundation for a theoretically-principled study of reinforcement learning under reward misspecification.

arXiv:2310.09123

Automatic Music Playlist Generation via Simulation-based Reinforcement Learning

Personalization of playlists is a common feature in music streaming services, but conventional techniques, such as collaborative filtering, rely on explicit assumptions regarding content quality to learn how to make recommendations. Such assumptions often result in misalignment between offline model objectives and online user satisfaction metrics. In this paper, we present a reinforcement learning framework that solves for such limitations by directly optimizing for user satisfaction metrics via the use of a simulated playlist-generation environment. Using this simulator we develop and train a modified Deep Q-Network, the action head DQN (AH-DQN), in a manner that addresses the challenges imposed by the large state and action space of our RL formulation. The resulting policy is capable of making recommendations from large and dynamic sets of candidate items with the expectation of maximizing consumption metrics. We analyze and evaluate agents offline via simulations that use environment models trained on both public and proprietary streaming datasets. We show how these agents lead to better user-satisfaction metrics compared to baseline methods during online A/B tests. Finally, we demonstrate that performance assessments produced from our simulator are strongly correlated with observed online metric results.

arXiv:2310.08731

Novelty Detection in Reinforcement Learning with World Models

Reinforcement learning (RL) using world models has found significant recent successes. However, when a sudden change to world mechanics or properties occurs then agent performance and reliability can dramatically decline. We refer to the sudden change in visual properties or state transitions as novelties. Implementing novelty detection within generated world model frameworks is a crucial task for protecting the agent when deployed. In this paper, we propose straightforward bounding approaches to incorporate novelty detection into world model RL agents, by utilizing the misalignment of the world model's hallucinated states and the true observed states as an anomaly score. We provide effective approaches to detecting novelties in a distribution of transitions learned by an agent in a world model. Finally, we show the advantage of our work in a novel environment compared to traditional machine learning novelty detection methods as well as currently accepted RL focused novelty detection algorithms.

arXiv:2310.08683공식 코드 · Papers with Code

Virtual Augmented Reality for Atari Reinforcement Learning

Reinforcement Learning (RL) has achieved significant milestones in the gaming domain, most notably Google DeepMind's AlphaGo defeating human Go champion Ken Jie. This victory was also made possible through the Atari Learning Environment (ALE): The ALE has been foundational in RL research, facilitating significant RL algorithm developments such as AlphaGo and others. In current Atari video game RL research, RL agents' perceptions of its environment is based on raw pixel data from the Atari video game screen with minimal image preprocessing. Contrarily, cutting-edge ML research, external to the Atari video game RL research domain, is focusing on enhancing image perception. A notable example is Meta Research's "Segment Anything Model" (SAM), a foundation model capable of segmenting images without prior training (zero-shot). This paper addresses a novel methodical question: Can state-of-the-art image segmentation models such as SAM improve the performance of RL agents playing Atari video games? The results suggest that SAM can serve as a "virtual augmented reality" for the RL agent, boosting its Atari video game playing performance under certain conditions. Comparing RL agent performance results from raw and augmented pixel inputs provides insight into these conditions. Although this paper was limited by computational constraints, the findings show improved RL agent performance for augmented pixel inputs and can inform broader research agendas in the domain of "virtual augmented reality for video game playing RL agents".

arXiv:2310.07745

Deep Reinforcement Learning for Autonomous Cyber Defence: A Survey

The rapid increase in the number of cyber-attacks in recent years raises the need for principled methods for defending networks against malicious actors. Deep reinforcement learning (DRL) has emerged as a promising approach for mitigating these attacks. However, while DRL has shown much potential for cyber defence, numerous challenges must be overcome before DRL can be applied to the autonomous cyber defence (ACD) problem at scale. Principled methods are required for environments that confront learners with very high-dimensional state spaces, large multi-discrete action spaces, and adversarial learning. Recent works have reported success in solving these problems individually. There have also been impressive engineering efforts towards solving all three for real-time strategy games. However, applying DRL to the full ACD problem remains an open challenge. Here, we survey the relevant DRL literature and conceptualize an idealised ACD-DRL agent. We provide: i.) A summary of the domain properties that define the ACD problem; ii.) A comprehensive comparison of current ACD environments used for benchmarking DRL approaches; iii.) An overview of state-of-the-art approaches for scaling DRL to domains that confront learners with the curse of dimensionality, and; iv.) A survey and critique of current methods for limiting the exploitability of agents within adversarial settings from the perspective of ACD. We conclude with open research questions that we hope will motivate future directions for researchers and practitioners working on ACD.

arXiv:2310.07981

Reinforcement Learning of Display Transfer Robots in Glass Flow Control Systems: A Physical Simulation-Based Approach

A flow control system is a critical concept for increasing the production capacity of manufacturing systems. To solve the scheduling optimization problem related to the flow control with the aim of improving productivity, existing methods depend on a heuristic design by domain human experts. Therefore, the methods require correction, monitoring, and verification by using real equipment. As system designs increase in complexity, the monitoring time increases, which decreases the probability of arriving at the optimal design. As an alternative approach to the heuristic design of flow control systems, the use of deep reinforcement learning to solve the scheduling optimization problem has been considered. Although the existing research on reinforcement learning has yielded excellent performance in some areas, the applicability of the results to actual FAB such as display and semiconductor manufacturing processes is not evident so far. To this end, we propose a method to implement a physical simulation environment and devise a feasible flow control system design using a transfer robot in display manufacturing through reinforcement learning. We present a model and parameter setting to build a virtual environment for different display transfer robots, and training methods of reinforcement learning on the environment to obtain an optimal scheduling of glass flow control systems. Its feasibility was verified by using different types of robots used in the actual process.

arXiv:2310.08387

Aligning Data Selection with Performance: Performance-driven Reinforcement Learning for Active Learning in Object Detection

Active learning strategies aim to train high-performance models with minimal labeled data by selecting the most informative instances for labeling. However, existing methods for assessing data informativeness often fail to align directly with task model performance metrics, such as mean average precision (mAP) in object detection. This paper introduces Mean-AP Guided Reinforced Active Learning for Object Detection (MGRAL), a novel approach that leverages the concept of expected model output changes as informativeness for deep detection networks, directly optimizing the sampling strategy using mAP. MGRAL employs a reinforcement learning agent based on LSTM architecture to efficiently navigate the combinatorial challenge of batch sample selection and the non-differentiable nature between performance and selected batches. The agent optimizes selection using policy gradient with mAP improvement as the reward signal. To address the computational intensity of mAP estimation with unlabeled samples, we implement fast look-up tables, ensuring real-world feasibility. We evaluate MGRAL on PASCAL VOC and MS COCO benchmarks across various backbone architectures. Our approach demonstrates strong performance, establishing a new paradigm in reinforcement learning-based active learning for object detection.

arXiv:2310.07899

RoboCLIP: One Demonstration is Enough to Learn Robot Policies

Reward specification is a notoriously difficult problem in reinforcement learning, requiring extensive expert supervision to design robust reward functions. Imitation learning (IL) methods attempt to circumvent these problems by utilizing expert demonstrations but typically require a large number of in-domain expert demonstrations. Inspired by advances in the field of Video-and-Language Models (VLMs), we present RoboCLIP, an online imitation learning method that uses a single demonstration (overcoming the large data requirement) in the form of a video demonstration or a textual description of the task to generate rewards without manual reward function design. Additionally, RoboCLIP can also utilize out-of-domain demonstrations, like videos of humans solving the task for reward generation, circumventing the need to have the same demonstration and deployment domains. RoboCLIP utilizes pretrained VLMs without any finetuning for reward generation. Reinforcement learning agents trained with RoboCLIP rewards demonstrate 2-3 times higher zero-shot performance than competing imitation learning methods on downstream robot manipulation tasks, doing so using only one video/text demonstration.

arXiv:2310.08566공식 코드 · Papers with Code

Transformers as Decision Makers: Provable In-Context Reinforcement Learning via Supervised Pretraining

Large transformer models pretrained on offline reinforcement learning datasets have demonstrated remarkable in-context reinforcement learning (ICRL) capabilities, where they can make good decisions when prompted with interaction trajectories from unseen environments. However, when and how transformers can be trained to perform ICRL have not been theoretically well-understood. In particular, it is unclear which reinforcement-learning algorithms transformers can perform in context, and how distribution mismatch in offline training data affects the learned algorithms. This paper provides a theoretical framework that analyzes supervised pretraining for ICRL. This includes two recently proposed training methods -- algorithm distillation and decision-pretrained transformers. First, assuming model realizability, we prove the supervised-pretrained transformer will imitate the conditional expectation of the expert algorithm given the observed trajectory. The generalization error will scale with model capacity and a distribution divergence factor between the expert and offline algorithms. Second, we show transformers with ReLU attention can efficiently approximate near-optimal online reinforcement learning algorithms like LinUCB and Thompson sampling for stochastic linear bandits, and UCB-VI for tabular Markov decision processes. This provides the first quantitative analysis of the ICRL capabilities of transformers pretrained from offline trajectories.

arXiv:2310.07613

Reinforcement Learning-based Knowledge Graph Reasoning for Explainable Fact-checking

Fact-checking is a crucial task as it ensures the prevention of misinformation. However, manual fact-checking cannot keep up with the rate at which false information is generated and disseminated online. Automated fact-checking by machines is significantly quicker than by humans. But for better trust and transparency of these automated systems, explainability in the fact-checking process is necessary. Fact-checking often entails contrasting a factual assertion with a body of knowledge for such explanations. An effective way of representing knowledge is the Knowledge Graph (KG). There have been sufficient works proposed related to fact-checking with the usage of KG but not much focus is given to the application of reinforcement learning (RL) in such cases. To mitigate this gap, we propose an RL-based KG reasoning approach for explainable fact-checking. Extensive experiments on FB15K-277 and NELL-995 datasets reveal that reasoning over a KG is an effective way of producing human-readable explanations in the form of paths and classifications for fact claims. The RL reasoning agent computes a path that either proves or disproves a factual claim, but does not provide a verdict itself. A verdict is reached by a voting mechanism that utilizes paths produced by the agent. These paths can be presented to human readers so that they themselves can decide whether or not the provided evidence is convincing or not. This work will encourage works in this direction for incorporating RL for explainable fact-checking as it increases trustworthiness by providing a human-in-the-loop approach.

arXiv:2310.06903

Reinforcement Learning in a Safety-Embedded MDP with Trajectory Optimization

Safe Reinforcement Learning (RL) plays an important role in applying RL algorithms to safety-critical real-world applications, addressing the trade-off between maximizing rewards and adhering to safety constraints. This work introduces a novel approach that combines RL with trajectory optimization to manage this trade-off effectively. Our approach embeds safety constraints within the action space of a modified Markov Decision Process (MDP). The RL agent produces a sequence of actions that are transformed into safe trajectories by a trajectory optimizer, thereby effectively ensuring safety and increasing training stability. This novel approach excels in its performance on challenging Safety Gym tasks, achieving significantly higher rewards and near-zero safety violations during inference. The method's real-world applicability is demonstrated through a safe and effective deployment in a real robot task of box-pushing around obstacles.

arXiv:2310.07433

Imitation Learning from Observation with Automatic Discount Scheduling

Humans often acquire new skills through observation and imitation. For robotic agents, learning from the plethora of unlabeled video demonstration data available on the Internet necessitates imitating the expert without access to its action, presenting a challenge known as Imitation Learning from Observations (ILfO). A common approach to tackle ILfO problems is to convert them into inverse reinforcement learning problems, utilizing a proxy reward computed from the agent's and the expert's observations. Nonetheless, we identify that tasks characterized by a progress dependency property pose significant challenges for such approaches; in these tasks, the agent needs to initially learn the expert's preceding behaviors before mastering the subsequent ones. Our investigation reveals that the main cause is that the reward signals assigned to later steps hinder the learning of initial behaviors. To address this challenge, we present a novel ILfO framework that enables the agent to master earlier behaviors before advancing to later ones. We introduce an Automatic Discount Scheduling (ADS) mechanism that adaptively alters the discount factor in reinforcement learning during the training phase, prioritizing earlier rewards initially and gradually engaging later rewards only when the earlier behaviors have been mastered. Our experiments, conducted on nine Meta-World tasks, demonstrate that our method significantly outperforms state-of-the-art methods across all tasks, including those that are unsolvable by them.

arXiv:2310.07207

Robust Safe Reinforcement Learning under Adversarial Disturbances

Safety is a primary concern when applying reinforcement learning to real-world control tasks, especially in the presence of external disturbances. However, existing safe reinforcement learning algorithms rarely account for external disturbances, limiting their applicability and robustness in practice. To address this challenge, this paper proposes a robust safe reinforcement learning framework that tackles worst-case disturbances. First, this paper presents a policy iteration scheme to solve for the robust invariant set, i.e., a subset of the safe set, where persistent safety is only possible for states within. The key idea is to establish a two-player zero-sum game by leveraging the safety value function in Hamilton-Jacobi reachability analysis, in which the protagonist (i.e., control inputs) aims to maintain safety and the adversary (i.e., external disturbances) tries to break down safety. This paper proves that the proposed policy iteration algorithm converges monotonically to the maximal robust invariant set. Second, this paper integrates the proposed policy iteration scheme into a constrained reinforcement learning algorithm that simultaneously synthesizes the robust invariant set and uses it for constrained policy optimization. This algorithm tackles both optimality and safety, i.e., learning a policy that attains high rewards while maintaining safety under worst-case disturbances. Experiments on classic control tasks show that the proposed method achieves zero constraint violation with learned worst-case adversarial disturbances, while other baseline algorithms violate the safety constraints substantially. Our proposed method also attains comparable performance as the baselines even in the absence of the adversary.

arXiv:2310.07518

Exploiting Causal Graph Priors with Posterior Sampling for Reinforcement Learning

Posterior sampling allows exploitation of prior knowledge on the environment's transition dynamics to improve the sample efficiency of reinforcement learning. The prior is typically specified as a class of parametric distributions, the design of which can be cumbersome in practice, often resulting in the choice of uninformative priors. In this work, we propose a novel posterior sampling approach in which the prior is given as a (partial) causal graph over the environment's variables. The latter is often more natural to design, such as listing known causal dependencies between biometric features in a medical treatment study. Specifically, we propose a hierarchical Bayesian procedure, called C-PSRL, simultaneously learning the full causal graph at the higher level and the parameters of the resulting factored dynamics at the lower level. We provide an analysis of the Bayesian regret of C-PSRL that explicitly connects the regret rate with the degree of prior knowledge. Our numerical evaluation conducted in illustrative domains confirms that C-PSRL strongly improves the efficiency of posterior sampling with an uninformative prior while performing close to posterior sampling with the full causal graph.