One of the most successful paradigms for reward learning uses human feedback in the form of comparisons. Although these methods hold promise, human comparison labeling is expensive and time consuming, constituting a major bottleneck to their broader applicability. Our insight is that we can greatly improve how effectively human time is used in these approaches by batching comparisons together, rather than having the human label each comparison individually. To do so, we leverage data dimensionality-reduction and visualization techniques to provide the human with a interactive GUI displaying the state space, in which the user can label subportions of the state space. Across some simple Mujoco tasks, we show that this high-level approach holds promise and is able to greatly increase the performance of the resulting agents, provided the same amount of human labeling time.
translated by 谷歌翻译
When robots learn reward functions using high capacity models that take raw state directly as input, they need to both learn a representation for what matters in the task -- the task ``features" -- as well as how to combine these features into a single objective. If they try to do both at once from input designed to teach the full reward function, it is easy to end up with a representation that contains spurious correlations in the data, which fails to generalize to new settings. Instead, our ultimate goal is to enable robots to identify and isolate the causal features that people actually care about and use when they represent states and behavior. Our idea is that we can tune into this representation by asking users what behaviors they consider similar: behaviors will be similar if the features that matter are similar, even if low-level behavior is different; conversely, behaviors will be different if even one of the features that matter differs. This, in turn, is what enables the robot to disambiguate between what needs to go into the representation versus what is spurious, as well as what aspects of behavior can be compressed together versus not. The notion of learning representations based on similarity has a nice parallel in contrastive learning, a self-supervised representation learning technique that maps visually similar data points to similar embeddings, where similarity is defined by a designer through data augmentation heuristics. By contrast, in order to learn the representations that people use, so we can learn their preferences and objectives, we use their definition of similarity. In simulation as well as in a user study, we show that learning through such similarity queries leads to representations that, while far from perfect, are indeed more generalizable than self-supervised and task-input alternatives.
translated by 谷歌翻译
While reinforcement learning (RL) has become a more popular approach for robotics, designing sufficiently informative reward functions for complex tasks has proven to be extremely difficult due their inability to capture human intent and policy exploitation. Preference based RL algorithms seek to overcome these challenges by directly learning reward functions from human feedback. Unfortunately, prior work either requires an unreasonable number of queries implausible for any human to answer or overly restricts the class of reward functions to guarantee the elicitation of the most informative queries, resulting in models that are insufficiently expressive for realistic robotics tasks. Contrary to most works that focus on query selection to \emph{minimize} the amount of data required for learning reward functions, we take an opposite approach: \emph{expanding} the pool of available data by viewing human-in-the-loop RL through the more flexible lens of multi-task learning. Motivated by the success of meta-learning, we pre-train preference models on prior task data and quickly adapt them for new tasks using only a handful of queries. Empirically, we reduce the amount of online feedback needed to train manipulation policies in Meta-World by 20$\times$, and demonstrate the effectiveness of our method on a real Franka Panda Robot. Moreover, this reduction in query-complexity allows us to train robot policies from actual human users. Videos of our results and code can be found at https://sites.google.com/view/few-shot-preference-rl/home.
translated by 谷歌翻译
当从人类行为中推断出奖励功能(无论是演示,比较,物理校正或电子停靠点)时,它已证明对人类进行建模作为做出嘈杂的理性选择,并具有“合理性系数”,以捕获多少噪声或熵我们希望看到人类的行为。无论人类反馈的类型或质量如何,许多现有作品都选择修复此系数。但是,在某些情况下,进行演示可能要比回答比较查询要困难得多。在这种情况下,我们应该期望在示范中看到比比较中更多的噪音或次级临时性,并且应该相应地解释反馈。在这项工作中,我们提倡,将每种反馈类型的实际数据中的理性系数扎根,而不是假设默认值,对奖励学习具有重大的积极影响。我们在模拟反馈以及用户研究的实验中测试了这一点。我们发现,从单一反馈类型中学习时,高估人类理性可能会对奖励准确性和遗憾产生可怕的影响。此外,我们发现合理性层面会影响每种反馈类型的信息性:令人惊讶的是,示威并不总是最有用的信息 - 当人类的行为非常卑鄙时,即使在合理性水平相同的情况下,比较实际上就变得更加有用。 。此外,当机器人确定要要求的反馈类型时,它可以通过准确建模每种类型的理性水平来获得很大的优势。最终,我们的结果强调了关注假定理性级别的重要性,不仅是在从单个反馈类型中学习时,尤其是当代理商从多种反馈类型中学习时,尤其是在学习时。
translated by 谷歌翻译
对于许多强化学习(RL)应用程序,指定奖励是困难的。本文考虑了一个RL设置,其中代理仅通过查询可以询问可以的专家来获取有关奖励的信息,例如,评估单个状态或通过轨迹提供二进制偏好。从如此昂贵的反馈中,我们的目标是学习奖励的模型,允许标准RL算法实现高预期的回报,尽可能少的专家查询。为此,我们提出了信息定向奖励学习(IDRL),它使用奖励的贝叶斯模型,然后选择要最大化信息增益的查询,这些查询是有关合理的最佳策略之间的返回差异的差异。与针对特定类型查询设计的先前主动奖励学习方法相比,IDRL自然地适应不同的查询类型。此外,它通过将焦点转移降低奖励近似误差来实现类似或更好的性能,从而降低奖励近似误差,以改善奖励模型引起的策略。我们支持我们的调查结果,在多个环境中进行广泛的评估,并具有不同的查询类型。
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
在嘈杂的互联网规模数据集上进行了预测,已对具有广泛的文本,图像和其他模式能力的培训模型进行了大量研究。但是,对于许多顺序决策域,例如机器人技术,视频游戏和计算机使用,公开可用的数据不包含以相同方式训练行为先验所需的标签。我们通过半监督的模仿学习将互联网规模的预处理扩展到顺序的决策域,其中代理通过观看在线未标记的视频来学习行动。具体而言,我们表明,使用少量标记的数据,我们可以训练一个足够准确的反向动力学模型,可以标记一个巨大的未标记在线数据来源 - 在这里,在线播放Minecraft的在线视频 - 然后我们可以从中训练一般行为先验。尽管使用了本地人类界面(鼠标和键盘为20Hz),但我们表明,这种行为先验具有非平凡的零射击功能,并且可以通过模仿学习和加强学习,可以对其进行微调,以进行硬探索任务。不可能通过增强学习从头开始学习。对于许多任务,我们的模型都表现出人类水平的性能,我们是第一个报告可以制作钻石工具的计算机代理,这些工具可以花费超过20分钟(24,000个环境动作)的游戏玩法来实现。
translated by 谷歌翻译
人类可以利用身体互动来教机器人武器。这种物理互动取决于任务,用户以及机器人到目前为止所学的内容。最先进的方法专注于从单一模态学习,或者假设机器人具有有关人类预期任务的先前信息,从而结合了多个互动类型。相比之下,在本文中,我们介绍了一种算法形式主义,该算法从演示,更正和偏好中学习。我们的方法对人类想要教机器人的任务没有任何假设。取而代之的是,我们通过将人类的输入与附近的替代方案进行比较,从头开始学习奖励模型。我们首先得出损失函数,该功能训练奖励模型的合奏,以匹配人类的示范,更正和偏好。反馈的类型和顺序取决于人类老师:我们使机器人能够被动地或积极地收集此反馈。然后,我们应用受约束的优化将我们学习的奖励转换为所需的机器人轨迹。通过模拟和用户研究,我们证明,与现有基线相比,我们提出的方法更准确地从人体互动中学习了操纵任务,尤其是当机器人面临新的或意外的目标时。我们的用户研究视频可在以下网址获得:https://youtu.be/fsujstyveku
translated by 谷歌翻译
我们研究了从机器人交互的大型离线数据集学习一系列基于视觉的操纵任务的问题。为了实现这一目标,人类需要简单有效地将任务指定给机器人。目标图像是一种流行的任务规范形式,因为它们已经在机器人的观察空间接地。然而,目标图像也有许多缺点:它们对人类提供的不方便,它们可以通过提供导致稀疏奖励信号的所需行为,或者在非目标达到任务的情况下指定任务信息。自然语言为任务规范提供了一种方便而灵活的替代方案,而是随着机器人观察空间的接地语言挑战。为了可扩展地学习此基础,我们建议利用具有人群源语言标签的离线机器人数据集(包括高度最佳,自主收集的数据)。使用此数据,我们学习一个简单的分类器,该分类器预测状态的更改是否完成了语言指令。这提供了一种语言调节奖励函数,然后可以用于离线多任务RL。在我们的实验中,我们发现,在语言条件的操作任务中,我们的方法优于目标 - 图像规格和语言条件仿制技术超过25%,并且能够从自然语言中执行Visuomotor任务,例如“打开右抽屉“和”移动订书机“,在弗兰卡·埃米卡熊猫机器人上。
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
我们调查视觉跨实施的模仿设置,其中代理商学习来自其他代理的视频(例如人类)的策略,示范相同的任务,但在其实施例中具有缺点差异 - 形状,动作,终效应器动态等。在这项工作中,我们证明可以从对这些差异强大的跨实施例证视频自动发现和学习基于视觉的奖励功能。具体而言,我们介绍了一种用于跨实施的跨实施的自我监督方法(XIRL),它利用时间周期 - 一致性约束来学习深度视觉嵌入,从而从多个专家代理的示范的脱机视频中捕获任务进度,每个都执行相同的任务不同的原因是实施例差异。在我们的工作之前,从自我监督嵌入产生奖励通常需要与参考轨迹对齐,这可能难以根据STARK实施例的差异来获取。我们凭经验显示,如果嵌入式了解任务进度,则只需在学习的嵌入空间中占据当前状态和目标状态之间的负距离是有用的,作为培训与加强学习的培训政策的奖励。我们发现我们的学习奖励功能不仅适用于在训练期间看到的实施例,而且还概括为完全新的实施例。此外,在将现实世界的人类示范转移到模拟机器人时,我们发现XIRL比当前最佳方法更具样本。 https://x-irl.github.io提供定性结果,代码和数据集
translated by 谷歌翻译
从演示和成对偏好推断奖励功能是对准与人类意图的强化学习(RL)代理的吉祥方法。然而,最先进的方法通常专注于学习单一奖励模型,从而使得难以从多个专家兑换不同的奖励功能。我们提出了多目标加强主动学习(道德),这是一种将社会规范多样化示范与帕累托最优政策相结合的新方法。通过维持分布在标量化权重,我们的方法能够以各种偏好交互地调整深度RL代理,同时消除了计算多个策略的需求。我们经验展示了道德在两种情景中的有效性,该方案模拟了需要代理人在规范冲突的情况下采取行动的交付和紧急任务。总体而言,我们认为我们的研究迈出了多目标RL的一步,具有学习奖励,弥合当前奖励学习和机器伦理文学之间的差距。
translated by 谷歌翻译
我们如何才能训练辅助人机接口(例如,基于肌电图的肢体假体),将用户的原始命令信号转换为机器人或计算机的动作,如果没有事先映射,我们不能要求用户进行监督动作标签或奖励反馈的形式,我们对用户试图完成的任务没有事先了解?本文中的关键想法是,无论任务如何,当接口更直观时,用户的命令就会不那么嘈杂。我们将这一想法形式化为一个完全无监督的目标,以优化接口:用户的命令信号与环境中的诱导状态过渡之间的相互信息。为了评估此相互信息得分是否可以区分有效的界面和无效界面,我们对540K的示例进行了观察性研究,该示例的用户操作各种键盘和眼睛凝视接口,用于打字,控制模拟机器人和玩视频游戏。结果表明,我们的共同信息得分可预测各个领域中的基础任务完成指标,而Spearman的平均等级相关性为0.43。除了对现有接口的离线评估外,我们还使用无监督的目标从头开始学习接口:我们随机初始化接口,让用户尝试使用接口执行其所需的任务,测量相互信息得分并更新接口通过强化学习最大化相互信息。我们通过用户研究与12名参与者进行用户研究评估我们的方法,他们使用扰动的鼠标执行2D光标控制任务,并使用手势使用手势的一个用户玩《 Lunar Lander》游戏的实验。结果表明,我们可以在30分钟内从头开始学习一个接头,无需任何用户监督或任务的先验知识。
translated by 谷歌翻译
深度加强学习(DEEPRL)方法已广泛用于机器人学,以了解环境,自主获取行为。深度互动强化学习(Deepirl)包括来自外部培训师或专家的互动反馈,提供建议,帮助学习者选择采取行动以加快学习过程。但是,目前的研究仅限于仅为特工现任提供可操作建议的互动。另外,在单个使用之后,代理丢弃该信息,该用途在为Revisit以相同状态引起重复过程。在本文中,我们提出了广泛的建议(BPA),这是一种广泛的持久的咨询方法,可以保留并重新使用加工信息。它不仅可以帮助培训师提供与类似状态相关的更一般性建议,而不是仅仅是当前状态,而且还允许代理加快学习过程。我们在两个连续机器人场景中测试提出的方法,即购物车极衡任务和模拟机器人导航任务。所得结果表明,使用BPA的代理的性能在于与深层方法相比保持培训师所需的相互作用的数量。
translated by 谷歌翻译
强化学习(RL)和脑电脑接口(BCI)是过去十年一直在增长的两个领域。直到最近,这些字段彼此独立操作。随着对循环(HITL)应用的兴趣升高,RL算法已经适用于人类指导,从而产生互动强化学习(IRL)的子领域。相邻的,BCI应用一直很感兴趣在人机交互期间从神经活动中提取内在反馈。这两个想法通过将BCI集成到IRL框架中,将RL和BCI设置在碰撞过程中,通过将内在反馈可用于帮助培训代理商来帮助框架。这种交叉点被称为内在的IRL。为了进一步帮助,促进BCI和IRL的更深层次,我们对内在IRILL的审查有着重点在于其母体领域的反馈驱动的IRL,同时还提供有关有效性,挑战和未来研究方向的讨论。
translated by 谷歌翻译
尽管深度强化学习(RL)最近取得了许多成功,但其方法仍然效率低下,这使得在数据方面解决了昂贵的许多问题。我们的目标是通过利用未标记的数据中的丰富监督信号来进行学习状态表示,以解决这一问题。本文介绍了三种不同的表示算法,可以访问传统RL算法使用的数据源的不同子集使用:(i)GRICA受到独立组件分析(ICA)的启发,并训练深层神经网络以输出统计独立的独立特征。输入。 Grica通过最大程度地减少每个功能与其他功能之间的相互信息来做到这一点。此外,格里卡仅需要未分类的环境状态。 (ii)潜在表示预测(LARP)还需要更多的上下文:除了要求状态作为输入外,它还需要先前的状态和连接它们的动作。该方法通过预测当前状态和行动的环境的下一个状态来学习状态表示。预测器与图形搜索算法一起使用。 (iii)重新培训通过训练深层神经网络来学习国家表示,以学习奖励功能的平滑版本。该表示形式用于预处理输入到深度RL,而奖励预测指标用于奖励成型。此方法仅需要环境中的状态奖励对学习表示表示。我们发现,每种方法都有其优势和缺点,并从我们的实验中得出结论,包括无监督的代表性学习在RL解决问题的管道中可以加快学习的速度。
translated by 谷歌翻译
强化学习(RL)需要访问刺激行为正确的行为的奖励功能,但这些都是非常难以指定复杂的任务。基于偏好RL提供了一种替代方案:用学习老师的偏好,而不用预先定义的奖励,从而克服与奖赏有关的工程关注的政策。然而,这是很难量化基于偏好-RL的进展,由于缺乏一个普遍采用的基准。在本文中,我们介绍了B-县:基准专为基于偏好-RL设计。这样的标杆的一个关键挑战是提供快速评估候选算法的能力,这使得依靠真正的人类输入的评价望而却步。与此同时,人类模拟输入作为给完美的喜好地面实况奖励功能是不现实的。 B-县通过一系列广泛的非理性的模拟教师缓解这一,并提出不仅仅是性能也为稳健性这些潜在的不合理性指标。我们用它来分析算法的设计选择,如选择信息查询,为国家的最先进的基于偏好的RL算法展示B-县的效用。我们希望B-县可以作为起点,以诚为本偏好研究RL更系统常见的。源代码可以在https://github.com/rll-research/B-Pref。
translated by 谷歌翻译
In the learning from demonstration (LfD) paradigm, understanding and evaluating the demonstrated behaviors plays a critical role in extracting control policies for robots. Without this knowledge, a robot may infer incorrect reward functions that lead to undesirable or unsafe control policies. Recent work has proposed an LfD framework where a user provides a set of formal task specifications to guide LfD, to address the challenge of reward shaping. However, in this framework, specifications are manually ordered in a performance graph (a partial order that specifies relative importance between the specifications). The main contribution of this paper is an algorithm to learn the performance graph directly from the user-provided demonstrations, and show that the reward functions generated using the learned performance graph generate similar policies to those from manually specified performance graphs. We perform a user study that shows that priorities specified by users on behaviors in a simulated highway driving domain match the automatically inferred performance graph. This establishes that we can accurately evaluate user demonstrations with respect to task specifications without expert criteria.
translated by 谷歌翻译
本文介绍了寻求信息(是)任务,概念和算法的信息重新分类。拟议的分类系统提供了新的维度,以研究寻求任务和方法的信息。新尺寸包括搜索迭代,搜索目标类型和程序的数量,以实现这些目标。寻求任务的信息沿着这些尺寸呼叫合适的计算解决方案的差异。然后,该文章评论了符合每个新类别的机器学习解决方案。该论文结束了对系统的评估活动进行了审查。
translated by 谷歌翻译
When robots interact with humans in homes, roads, or factories the human's behavior often changes in response to the robot. Non-stationary humans are challenging for robot learners: actions the robot has learned to coordinate with the original human may fail after the human adapts to the robot. In this paper we introduce an algorithmic formalism that enables robots (i.e., ego agents) to co-adapt alongside dynamic humans (i.e., other agents) using only the robot's low-level states, actions, and rewards. A core challenge is that humans not only react to the robot's behavior, but the way in which humans react inevitably changes both over time and between users. To deal with this challenge, our insight is that -- instead of building an exact model of the human -- robots can learn and reason over high-level representations of the human's policy and policy dynamics. Applying this insight we develop RILI: Robustly Influencing Latent Intent. RILI first embeds low-level robot observations into predictions of the human's latent strategy and strategy dynamics. Next, RILI harnesses these predictions to select actions that influence the adaptive human towards advantageous, high reward behaviors over repeated interactions. We demonstrate that -- given RILI's measured performance with users sampled from an underlying distribution -- we can probabilistically bound RILI's expected performance across new humans sampled from the same distribution. Our simulated experiments compare RILI to state-of-the-art representation and reinforcement learning baselines, and show that RILI better learns to coordinate with imperfect, noisy, and time-varying agents. Finally, we conduct two user studies where RILI co-adapts alongside actual humans in a game of tag and a tower-building task. See videos of our user studies here: https://youtu.be/WYGO5amDXbQ
translated by 谷歌翻译