Dealing with sparse rewards is one of the biggest challenges in Reinforcement Learning (RL). We present a novel technique called Hindsight Experience Replay which allows sample-efficient learning from rewards which are sparse and binary and therefore avoid the need for complicated reward engineering. It can be combined with an arbitrary off-policy RL algorithm and may be seen as a form of implicit curriculum. We demonstrate our approach on the task of manipulating objects with a robotic arm. In particular, we run experiments on three different tasks: pushing, sliding, and pick-and-place, in each case using only binary rewards indicating whether or not the task is completed. Our ablation studies show that Hindsight Experience Replay is a crucial ingredient which makes training possible in these challenging environments. We show that our policies trained on a physics simulation can be deployed on a physical robot and successfully complete the task. The video presenting our experiments is available at https://goo.gl/SMrQnI.
translated by 谷歌翻译
Exploration in environments with sparse rewards has been a persistent problem in reinforcement learning (RL). Many tasks are natural to specify with a sparse reward, and manually shaping a reward function can result in suboptimal performance. However, finding a non-zero reward is exponentially more difficult with increasing task horizon or action dimensionality. This puts many real-world tasks out of practical reach of RL methods. In this work, we use demonstrations to overcome the exploration problem and successfully learn to perform long-horizon, multi-step robotics tasks with continuous control such as stacking blocks with a robot arm. Our method, which builds on top of Deep Deterministic Policy Gradients and Hindsight Experience Replay, provides an order of magnitude of speedup over RL on simulated robotics tasks. It is simple to implement and makes only the additional assumption that we can collect a small set of demonstrations. Furthermore, our method is able to solve tasks not solvable by either RL or behavior cloning alone, and often ends up outperforming the demonstrator policy.
translated by 谷歌翻译
通过加强学习(RL)掌握机器人操纵技巧通常需要设计奖励功能。该地区的最新进展表明,使用稀疏奖励,即仅在成功完成任务时奖励代理,可能会导致更好的政策。但是,在这种情况下,国家行动空间探索更困难。最近的RL与稀疏奖励学习的方法已经为任务提供了高质量的人类演示,但这些可能是昂贵的,耗时甚至不可能获得的。在本文中,我们提出了一种不需要人类示范的新颖有效方法。我们观察到,每个机器人操纵任务都可以被视为涉及从被操纵对象的角度来看运动的任务,即,对象可以了解如何自己达到目标状态。为了利用这个想法,我们介绍了一个框架,最初使用现实物理模拟器获得对象运动策略。然后,此策略用于生成辅助奖励,称为模拟的机器人演示奖励(SLDRS),使我们能够学习机器人操纵策略。拟议的方法已在增加复杂性的13个任务中进行了评估,与替代算法相比,可以实现更高的成功率和更快的学习率。 SLDRS对多对象堆叠和非刚性物体操作等任务特别有益。
translated by 谷歌翻译
We adapt the ideas underlying the success of Deep Q-Learning to the continuous action domain. We present an actor-critic, model-free algorithm based on the deterministic policy gradient that can operate over continuous action spaces. Using the same learning algorithm, network architecture and hyper-parameters, our algorithm robustly solves more than 20 simulated physics tasks, including classic problems such as cartpole swing-up, dexterous manipulation, legged locomotion and car driving. Our algorithm is able to find policies whose performance is competitive with those found by a planning algorithm with full access to the dynamics of the domain and its derivatives. We further demonstrate that for many of the tasks the algorithm can learn policies "end-to-end": directly from raw pixel inputs.
translated by 谷歌翻译
通过稀疏奖励的环境中的深度加强学习学习机器人操纵是一项具有挑战性的任务。在本文中,我们通过引入虚构对象目标的概念来解决这个问题。对于给定的操纵任务,首先通过物理逼真的模拟训练感兴趣的对象以达到自己的目标位置,而不会被操纵。然后利用对象策略来构建可编征物体轨迹的预测模型,该轨迹提供具有逐步更加困难的对象目标的机器人来达到训练期间的课程。所提出的算法,遵循对象(FO),已经在需要增加探索程度的7个Mujoco环境中进行评估,并且与替代算法相比,取得了更高的成功率。在特别具有挑战性的学习场景中,例如当物体的初始和目标位置相隔甚远,我们的方法仍然可以学习政策,而竞争方法目前失败。
translated by 谷歌翻译
多目标增强学习被广泛应用于计划和机器人操纵中。多进球强化学习的两个主要挑战是稀疏的奖励和样本效率低下。 Hindsight Experience重播(她)旨在通过进球重新标记来应对这两个挑战。但是,与她相关的作品仍然需要数百万个样本和庞大的计算。在本文中,我们提出了多步事化经验重播(MHER),并根据$ n $ step Relabeling合并了多步重新标记的回报,以提高样品效率。尽管$ n $ step Relableling具有优势,但我们从理论上和实验上证明了$ n $ step Relabeling引入的非政策$ n $步骤偏置可能会导致许多环境的性能差。为了解决上述问题,提出了两种偏差降低的MHER算法,Mher($ \ lambda $)和基于模型的Mher(Mmher)。 Mher($ \ lambda $)利用$ \ lambda $返回,而Mmher从基于模型的价值扩展中受益。对众多多目标机器人任务的实验结果表明,我们的解决方案可以成功减轻$ n $ n $步骤的偏见,并获得比她的样本效率明显更高,并且课程引导她,而她几乎没有其他计算。
translated by 谷歌翻译
后敏感经验重播(她)是一种常见的脱离政策深度加强学习算法的目标,以解决面向目标的任务;它非常适合提供仅提供稀疏奖励的机器人操纵任务。在她身上,轨迹和过渡都是均匀地对训练进行采样的。然而,并非所有的代理商的经历都同样促进训练,因此天真的统一采样可能导致学习效率低下。在本文中,我们提出了与她(DTGSH)的多样性轨迹和目标选择。首先,根据目标状态的多样性对由决定点过程(DPP)的模型进行采样进行采样。其次,通过使用K-DPP从轨迹中选择具有不同目标状态的转换。我们在模拟机器人环境中评估五个挑战机器人操纵任务的DTGSH,在那里我们表明我们的方法可以更快地学到更快,并且比所有任务的其他最先进的方法达到更高的性能。
translated by 谷歌翻译
在环境中的多进球强化学习中,代理商通过利用从与环境的互动中获得的经验来学习实现多个目标的政策。由于缺乏成功的经验,培训代理人凭借稀疏的二元奖励特别具有挑战性。为了解决这个问题,事后观察体验重播(她)从失败的经历中获得了成功的经验。但是,在不考虑实现目标财产的情况下产生成功的经验效率较低。在本文中,提出了一种基于集群的采样策略,利用实现目标的财产。提出的采样策略小组以不同的方式实现了目标和样本经历。对于分组,使用K-均值聚类算法。集群的质心是从定义为未实现的原始目标的失败目标的分布中获得的。该方法通过使用OpenAI健身房的三个机器人控制任务进行实验来验证。实验的结果表明,所提出的方法显着减少了在这三个任务中的两个中收敛所需的时期数量,并略微增加了其余一个任务的成功率。还表明,提出的方法可以与她的其他抽样策略结合使用。
translated by 谷歌翻译
强化学习的许多应用都可以正式化为目标条件的环境,在每个情节中,都有一个“目标”会影响该情节中获得的奖励,但不会影响动态。已经提出了各种技术来提高目标条件环境的性能,例如自动课程生成和目标重新标记。在这项工作中,我们探讨了在目标条件设置中的损失钢筋学习与知识蒸馏之间的联系。特别是:当前的Q值函数和目标Q值估计是该目标的函数,我们想训练Q值函数以匹配其所有目标的目标。因此,我们将基于梯度的注意转移(Zagoruyko和Komodakis 2017)(一种知识蒸馏技术)应用于Q功能更新。我们从经验上表明,当目标空间高维时,这可以提高目标条件的非政策强化学习的性能。我们还表明,在多个同时稀疏目标的情况下,可以对该技术进行调整,以允许有效学习,在这种情况下,代理可以通过在测试时间指定的所有大型目标来实现奖励。最后,为了提供理论支持,我们给出了环境类别的示例,在某些假设下(在某些假设)中,标准的非政策算法至少需要O(d^2)观察到的过渡以学习最佳策略,而我们的建议技术仅需O( d)过渡,其中d是目标和状态空间的维度。
translated by 谷歌翻译
强化学习是机器人抓握的一种有前途的方法,因为它可以在困难的情况下学习有效的掌握和掌握政策。但是,由于问题的高维度,用精致的机器人手来实现类似人类的操纵能力是具有挑战性的。尽管可以采用奖励成型或专家示范等补救措施来克服这个问题,但它们通常导致过分简化和有偏见的政策。我们介绍了Dext-Gen,这是一种在稀疏奖励环境中灵巧抓握的强化学习框架,适用于各种抓手,并学习无偏见和复杂的政策。通过平滑方向表示实现了抓地力和物体的完全方向控制。我们的方法具有合理的培训时间,并提供了包括所需先验知识的选项。模拟实验证明了框架对不同方案的有效性和适应性。
translated by 谷歌翻译
In order to avoid conventional controlling methods which created obstacles due to the complexity of systems and intense demand on data density, developing modern and more efficient control methods are required. In this way, reinforcement learning off-policy and model-free algorithms help to avoid working with complex models. In terms of speed and accuracy, they become prominent methods because the algorithms use their past experience to learn the optimal policies. In this study, three reinforcement learning algorithms; DDPG, TD3 and SAC have been used to train Fetch robotic manipulator for four different tasks in MuJoCo simulation environment. All of these algorithms are off-policy and able to achieve their desired target by optimizing both policy and value functions. In the current study, the efficiency and the speed of these three algorithms are analyzed in a controlled environment.
translated by 谷歌翻译
Meta-Renifiltive学习(Meta-RL)已被证明是利用事先任务的经验,以便快速学习新的相关任务的成功框架,但是,当前的Meta-RL接近在稀疏奖励环境中学习的斗争。尽管现有的Meta-RL算法可以学习适应新的稀疏奖励任务的策略,但是使用手形奖励功能来学习实际适应策略,或者需要简单的环境,其中随机探索足以遇到稀疏奖励。在本文中,我们提出了对Meta-RL的后视抢购的制定,该rl抢购了在Meta培训期间的经验,以便能够使用稀疏奖励完全学习。我们展示了我们的方法在套件挑战稀疏奖励目标达到的环境中,以前需要密集的奖励,以便在Meta训练中解决。我们的方法使用真正的稀疏奖励功能来解决这些环境,性能与具有代理密集奖励功能的培训相当。
translated by 谷歌翻译
尽管深入的强化学习(DRL)在包括机器人技术在内的许多学科中都很流行,但最先进的DRL算法仍然难以学习长途,多步骤和稀疏奖励任务,例如仅在只有一项任务的情况下堆叠几个块 - 集合奖励信号。为了提高此类任务的学习效率,本文提出了一种称为A^2的DRL探索技术,该技术集成了受人类经验启发的两个组成部分:抽象演示和适应性探索。 A^2首先将复杂的任务分解为子任务,然后提供正确的子任务订单以学习。在训练过程中,该代理商会自适应地探索环境,对良好的子任务的行为更确定性,并且更随机地对不良的子任务子任务。消融和比较实验是对几个网格世界任务和三个机器人操纵任务进行的。我们证明A^2可以帮助流行的DRL算法(DQN,DDPG和SAC)在这些环境中更有效,稳定地学习。
translated by 谷歌翻译
For an autonomous agent to fulfill a wide range of user-specified goals at test time, it must be able to learn broadly applicable and general-purpose skill repertoires. Furthermore, to provide the requisite level of generality, these skills must handle raw sensory input such as images. In this paper, we propose an algorithm that acquires such general-purpose skills by combining unsupervised representation learning and reinforcement learning of goal-conditioned policies. Since the particular goals that might be required at test-time are not known in advance, the agent performs a self-supervised "practice" phase where it imagines goals and attempts to achieve them. We learn a visual representation with three distinct purposes: sampling goals for self-supervised practice, providing a structured transformation of raw sensory inputs, and computing a reward signal for goal reaching. We also propose a retroactive goal relabeling scheme to further improve the sample-efficiency of our method. Our off-policy algorithm is efficient enough to learn policies that operate on raw image observations and goals for a real-world robotic system, and substantially outperforms prior techniques. * Equal contribution. Order was determined by coin flip.
translated by 谷歌翻译
本文详细介绍了我们对2021年真正机器人挑战的第一阶段提交的提交;三指机器人必须沿指定目标轨迹携带立方体的挑战。为了解决第1阶段,我们使用一种纯净的增强学习方法,该方法需要对机器人系统或机器人抓握的最少专家知识。与事后的经验重播一起采用了稀疏,基于目标的奖励,以教导控制立方体将立方体移至目标的X和Y坐标。同时,采用了基于密集的距离奖励来教授将立方体提升到目标的Z坐标(高度组成部分)的政策。该策略在将域随机化的模拟中进行培训,然后再转移到真实的机器人进行评估。尽管此次转移后的性能往往会恶化,但我们的最佳政策可以通过有效的捏合掌握能够成功地沿目标轨迹提升真正的立方体。我们的方法表现优于所有其他提交,包括那些利用更传统的机器人控制技术的提交,并且是第一个解决这一挑战的纯学习方法。
translated by 谷歌翻译
深度加固学习(DRL)使机器人能够结束结束地执行一些智能任务。然而,长地平线稀疏奖励机器人机械手任务仍存在许多挑战。一方面,稀疏奖励设置会导致探索效率低下。另一方面,使用物理机器人的探索是高成本和不安全的。在本文中,我们提出了一种学习使用本文中名为基础控制器的一个或多个现有传统控制器的长地平线稀疏奖励任务。基于深度确定性的政策梯度(DDPG),我们的算法将现有基础控制器融入勘探,价值学习和策略更新的阶段。此外,我们介绍了合成不同基础控制器以整合它们的优点的直接方式。通过从堆叠块到杯子的实验,证明学习的国家或基于图像的策略稳定优于基础控制器。与以前的示范中的学习作品相比,我们的方法通过数量级提高了样品效率,提高了性能。总体而言,我们的方法具有利用现有的工业机器人操纵系统来构建更灵活和智能控制器的可能性。
translated by 谷歌翻译
Reinforcement learning holds the promise of enabling autonomous robots to learn large repertoires of behavioral skills with minimal human intervention. However, robotic applications of reinforcement learning often compromise the autonomy of the learning process in favor of achieving training times that are practical for real physical systems. This typically involves introducing hand-engineered policy representations and human-supplied demonstrations. Deep reinforcement learning alleviates this limitation by training general-purpose neural network policies, but applications of direct deep reinforcement learning algorithms have so far been restricted to simulated settings and relatively simple tasks, due to their apparent high sample complexity. In this paper, we demonstrate that a recent deep reinforcement learning algorithm based on offpolicy training of deep Q-functions can scale to complex 3D manipulation tasks and can learn deep neural network policies efficiently enough to train on real physical robots. We demonstrate that the training times can be further reduced by parallelizing the algorithm across multiple robots which pool their policy updates asynchronously. Our experimental evaluation shows that our method can learn a variety of 3D manipulation skills in simulation and a complex door opening skill on real robots without any prior demonstrations or manually designed representations.
translated by 谷歌翻译
Reinforcement learning in partially observable domains is challenging due to the lack of observable state information. Thankfully, learning offline in a simulator with such state information is often possible. In particular, we propose a method for partially observable reinforcement learning that uses a fully observable policy (which we call a state expert) during offline training to improve online performance. Based on Soft Actor-Critic (SAC), our agent balances performing actions similar to the state expert and getting high returns under partial observability. Our approach can leverage the fully-observable policy for exploration and parts of the domain that are fully observable while still being able to learn under partial observability. On six robotics domains, our method outperforms pure imitation, pure reinforcement learning, the sequential or parallel combination of both types, and a recent state-of-the-art method in the same setting. A successful policy transfer to a physical robot in a manipulation task from pixels shows our approach's practicality in learning interesting policies under partial observability.
translated by 谷歌翻译
稀疏奖励学习通常在加强学习(RL)方面效率低下。 Hindsight Experience重播(她)已显示出一种有效的解决方案,可以处理低样本效率,这是由于目标重新标记而导致的稀疏奖励效率。但是,她仍然有一个隐含的虚拟阳性稀疏奖励问题,这是由于实现目标而引起的,尤其是对于机器人操纵任务而言。为了解决这个问题,我们提出了一种新型的无模型连续RL算法,称为Relay-HER(RHER)。提出的方法首先分解并重新布置原始的长马任务,以增量复杂性为新的子任务。随后,多任务网络旨在以复杂性的上升顺序学习子任务。为了解决虚拟阳性的稀疏奖励问题,我们提出了一种随机混合的探索策略(RME),在该策略中,在复杂性较低的人的指导下,较高复杂性的子任务的实现目标很快就会改变。实验结果表明,在五个典型的机器人操纵任务中,与香草盖相比,RHER样品效率的显着提高,包括Push,Pickandplace,抽屉,插入物和InstaclePush。提出的RHER方法还应用于从头开始的物理机器人上的接触式推送任务,成功率仅使用250集达到10/10。
translated by 谷歌翻译
强化学习(RL)代理商可以通过与环境进行交互来学习解决复杂的顺序决策任务。但是,样品效率仍然是一个重大挑战。在多目标RL领域中,需要代理以达到多个目标来解决复杂任务,提高采样效率可能尤其具有挑战性。另一方面,人类或其他生物代理商以更具战略方式学习此类任务,遵循随着难度水平的增加,以便逐步高效的学习进步。在这项工作中,我们提出了一种以自我监督方式使用动态距离功能(DDF)的自动目标生成方法。 DDF是一种函数,它预测马尔可夫决策过程(MDP)内的任何两个状态之间的动态距离。有了这个,我们在适当的难度水平下生成一个目标课程,以便在整个培训过程中有效地学习。我们在几个目标条件的机器人操纵和导航任务中评估这种方法,并在基线方法上显示出样本效率的改进,该方法仅使用随机目标采样。
translated by 谷歌翻译