In this paper we examine the problem of determining demonstration sufficiency for AI agents that learn from demonstrations: how can an AI agent self-assess whether it has received enough demonstrations from an expert to ensure a desired level of performance? To address this problem we propose a novel self-assessment approach based on Bayesian inverse reinforcement learning and value-at-risk to enable agents that learn from demonstrations to compute high-confidence bounds on their performance and use these bounds to determine when they have a sufficient number of demonstrations. We propose and evaluate two definitions of sufficiency: (1) normalized expected value difference, which measures regret with respect to the expert's unobserved reward function, and (2) improvement over a baseline policy. We demonstrate how to formulate high-confidence bounds on both of these metrics. We evaluate our approach in simulation and demonstrate the feasibility of developing an AI system that can accurately evaluate whether it has received sufficient training data to guarantee, with high confidence, that it can match an expert's performance or surpass the performance of a baseline policy within some desired safety threshold.
translated by 谷歌翻译
Inferring reward functions from human behavior is at the center of value alignment - aligning AI objectives with what we, humans, actually want. But doing so relies on models of how humans behave given their objectives. After decades of research in cognitive science, neuroscience, and behavioral economics, obtaining accurate human models remains an open research topic. This begs the question: how accurate do these models need to be in order for the reward inference to be accurate? On the one hand, if small errors in the model can lead to catastrophic error in inference, the entire framework of reward learning seems ill-fated, as we will never have perfect models of human behavior. On the other hand, if as our models improve, we can have a guarantee that reward accuracy also improves, this would show the benefit of more work on the modeling side. We study this question both theoretically and empirically. We do show that it is unfortunately possible to construct small adversarial biases in behavior that lead to arbitrarily large errors in the inferred reward. However, and arguably more importantly, we are also able to identify reasonable assumptions under which the reward inference error can be bounded linearly in the error in the human model. Finally, we verify our theoretical insights in discrete and continuous control tasks with simulated and human data.
translated by 谷歌翻译
Safe Reinforcement Learning can be defined as the process of learning policies that maximize the expectation of the return in problems in which it is important to ensure reasonable system performance and/or respect safety constraints during the learning and/or deployment processes. We categorize and analyze two approaches of Safe Reinforcement Learning. The first is based on the modification of the optimality criterion, the classic discounted finite/infinite horizon, with a safety factor. The second is based on the modification of the exploration process through the incorporation of external knowledge or the guidance of a risk metric. We use the proposed classification to survey the existing literature, as well as suggesting future directions for Safe Reinforcement Learning.
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
逆增强学习(IRL)是从专家演示中推断奖励功能的强大范式。许多IRL算法都需要已知的过渡模型,有时甚至是已知的专家政策,或者至少需要访问生成模型。但是,对于许多现实世界应用,这些假设太强了,在这些应用程序中,只能通过顺序相互作用访问环境。我们提出了一种新颖的IRL算法:逆增强学习(ACEIRL)的积极探索,该探索积极探索未知的环境和专家政策,以快速学习专家的奖励功能并确定良好的政策。 Aceirl使用以前的观察来构建置信区间,以捕获合理的奖励功能,并找到关注环境最有用区域的勘探政策。 Aceirl是使用样品复杂性界限的第一种活动IRL的方法,不需要环境的生成模型。在最坏情况下,Aceirl与活性IRL的样品复杂性与生成模型匹配。此外,我们建立了一个与问题相关的结合,该结合将Aceirl的样品复杂性与给定IRL问题的次级隔离间隙联系起来。我们在模拟中对Aceirl进行了经验评估,发现它的表现明显优于更幼稚的探索策略。
translated by 谷歌翻译
我们研究了设计AI代理商的问题,该代理可以学习有效地与潜在的次优伴侣有效合作,同时无法访问联合奖励功能。这个问题被建模为合作焦论双代理马尔可夫决策过程。我们假设仅在游戏的Stackelberg制定中的两个代理中的第一个控制,其中第二代理正在作用,以便在鉴于第一代理的政策给出预期的效用。第一个代理人应该如何尽快学习联合奖励功能,因此联合政策尽可能接近最佳?在本文中,我们分析了如何在这一交互式的两个代理方案中获得对奖励函数的知识。我们展示当学习代理的策略对转换函数有显着影响时,可以有效地学习奖励功能。
translated by 谷歌翻译
在本文中,我们研究了不确定性下的顺序决策任务中可读性的概念。以前的作品将易读性扩展到了机器人运动以外的方案,要么集中在确定性设置上,要么在计算上太昂贵。我们提出的称为POL-MDP的方法能够处理不确定性,同时保持计算障碍。在几种不同复杂性的模拟场景中,我们建立了反对最新方法的方法的优势。我们还展示了将我们的清晰政策用作反向加强学习代理的示范,并根据最佳政策建立了他们的优越性。最后,我们通过用户研究评估计算政策的可读性,在该研究中,要求人们通过观察其行动来推断移动机器人的目标。
translated by 谷歌翻译
从演示中学习的方法(LFD)通过模仿用户表现出在获取行为策略方面的成功。但是,即使对于一项任务,LFD也可能需要大量的演示。对于必须通过演示学习许多任务的多功能代理,如果孤立地学习每个任务,此过程将大大负担用户的负担。为了应对这一挑战,我们介绍了从演示中学习的新颖问题,该问题使代理商能够不断地基于从先前演示的任务中学到的知识,以加速学习新任务,从而减少所需的示范量。作为解决这个问题的一种解决方案,我们提出了第一种终身学习方法来进行逆强化学习,该方法通过演示学习连续的任务,不断地在任务之间转移知识以提高绩效。
translated by 谷歌翻译
真实世界的机器人任务需要复杂的奖励功能。当我们定义机器人需要解决的问题时,我们假装设计人员确切地指定了这种复杂的奖励,并且从那时起,它被设置为石头。然而,在实践中,奖励设计是一个迭代过程:设计师选择奖励,最终遇到奖励激励错误行为的“边缘案例”环境,修改奖励和重复。重新思考机器人问题是什么意思,正式占奖励设计的这种迭代性质?我们建议机器人不采取特定的奖励,而是对其进行不确定性,并占未来设计迭代作为未来的证据。我们贡献了辅助奖励设计方法,通过预测和影响未来的证据来加速设计过程:而不是让设计师最终遇到故障情况并修改奖励,该方法在开发阶段主动地将设计者暴露于这种环境。我们在简化的自主驾驶任务中测试此方法,并发现它通过提出当前奖励的“边缘案例”的环境,更快地提高汽车的行为。
translated by 谷歌翻译
当将强化学习(RL)代理部署到物理系统中时,我们必须确保这些代理非常了解基本的约束。但是,在许多现实世界中,遵循的限制因素(例如,人类)通常很难在数学上和RL代理商上指定。为了解决这些问题,约束逆强化学习(CIRL)考虑了约束马尔可夫决策过程(CMDP)的形式主义,并通过学习约束功能来估算专家示范中的约束。作为一个新兴的研究主题,Cirl没有共同的基准测试,以前的作品通过手工制作的环境(例如,网格世界)测试了其算法。在本文中,我们在两个主要的应用域:机器人控制和自动驾驶的背景下构建了CIRL基准。我们为每个环境设计相关的约束,并经验研究不同算法基于尊重这些约束的专家轨迹恢复这些约束的能力。为了处理随机动力学,我们提出了一种差异方法,以扩展约束分布,并通过将其与基准上的其他cirl基线进行比较来证明其性能。基准,包括复制CIRL算法性能的信息,可在https://github.com/guiliang/guiliang/cirl-benchmarks-public上公开获得
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
为了与机器人合作,我们必须能够理解他们的决策。人类自然会通过类似于逆增强学习(IRL)的方式来推理其可观察到的行为,从而推断出其他代理商的信念和欲望。因此,机器人可以通过提供对人类学习者的IRL提供信息的示威来传达他们的信念和欲望。一项内容丰富的演示是,鉴于他们当前对机器人决策的理解,与学习者对机器人将要做的事情的期望有很大差异。但是,标准IRL并未对学习者的现有期望进行建模,因此不能执行这种反事实推理。我们建议将学习者对机器人决策的当前理解纳入我们的人类IRL模型中,以便机器人可以选择最大化人类理解的演示。我们还提出了一种新颖的措施,以估计人类在看不见环境中预测机器人行为的实例的难度。一项用户研究发现,我们的测试难度与人类绩效和信心息息相关。有趣的是,选择人类的信念和反事实时,选择示范会在易于测试中降低人类绩效,但在困难测试中提高了性能,从而提供了有关如何最好地利用此类模型的见解。
translated by 谷歌翻译
In inverse reinforcement learning (IRL), a learning agent infers a reward function encoding the underlying task using demonstrations from experts. However, many existing IRL techniques make the often unrealistic assumption that the agent has access to full information about the environment. We remove this assumption by developing an algorithm for IRL in partially observable Markov decision processes (POMDPs). We address two limitations of existing IRL techniques. First, they require an excessive amount of data due to the information asymmetry between the expert and the learner. Second, most of these IRL techniques require solving the computationally intractable forward problem -- computing an optimal policy given a reward function -- in POMDPs. The developed algorithm reduces the information asymmetry while increasing the data efficiency by incorporating task specifications expressed in temporal logic into IRL. Such specifications may be interpreted as side information available to the learner a priori in addition to the demonstrations. Further, the algorithm avoids a common source of algorithmic complexity by building on causal entropy as the measure of the likelihood of the demonstrations as opposed to entropy. Nevertheless, the resulting problem is nonconvex due to the so-called forward problem. We solve the intrinsic nonconvexity of the forward problem in a scalable manner through a sequential linear programming scheme that guarantees to converge to a locally optimal policy. In a series of examples, including experiments in a high-fidelity Unity simulator, we demonstrate that even with a limited amount of data and POMDPs with tens of thousands of states, our algorithm learns reward functions and policies that satisfy the task while inducing similar behavior to the expert by leveraging the provided side information.
translated by 谷歌翻译
假设人类(大约)理性使机器人能够通过观察人类行为来推断奖励功能。但人们展出了各种各样的非理性,我们与这项工作的目标是更好地了解他们可以对奖励推论的影响。研究这种效果的挑战是存在许多类型的非理性,具有不同程度的数学形式化。因此,通过改变Bellman Optimaly公式,使用本框架来研究这些框架会如何影响推理的框架,从而通过改变MDP的语言。我们发现错误地建模一个系统地造型的人类,因为嘈杂的理性比正确捕获这些偏差更糟糕 - 这么多,因此可以更好地跳过推动并坚持先前!更重要的是,我们表明,在正确建模时,一个非理性人类可以传达有关奖励的更多信息,而不是完全合理的人体。也就是说,如果机器人具有正确的人类非理性模型,如果人类是理性的,它可以使推论比它能够更强大。非理性基本上有助于而不是阻碍奖励推断,但需要正确占用。
translated by 谷歌翻译
本文解决了逆增强学习(IRL)的问题 - 从观察其行为中推断出代理的奖励功能。 IRL可以为学徒学习提供可概括和紧凑的代表,并能够准确推断人的偏好以帮助他们。 %并提供更准确的预测。但是,有效的IRL具有挑战性,因为许多奖励功能可以与观察到的行为兼容。我们专注于如何利用先前的强化学习(RL)经验,以使学习这些偏好更快,更高效。我们提出了IRL算法基础(通过样本中的连续功能意图推断行为获取行为),该算法利用多任务RL预培训和后继功能,使代理商可以为跨越可能的目标建立强大的基础,从而跨越可能的目标。给定的域。当仅接触一些专家演示以优化新颖目标时,代理商会使用其基础快速有效地推断奖励功能。我们的实验表明,我们的方法非常有效地推断和优化显示出奖励功能,从而准确地从少于100个轨迹中推断出奖励功能。
translated by 谷歌翻译
Learning from Demonstration (LfD) is a powerful method for enabling robots to perform novel tasks as it is often more tractable for a non-roboticist end-user to demonstrate the desired skill and for the robot to efficiently learn from the associated data than for a human to engineer a reward function for the robot to learn the skill via reinforcement learning (RL). Safety issues arise in modern LfD techniques, e.g., Inverse Reinforcement Learning (IRL), just as they do for RL; yet, safe learning in LfD has received little attention. In the context of agile robots, safety is especially vital due to the possibility of robot-environment collision, robot-human collision, and damage to the robot. In this paper, we propose a safe IRL framework, CBFIRL, that leverages the Control Barrier Function (CBF) to enhance the safety of the IRL policy. The core idea of CBFIRL is to combine a loss function inspired by CBF requirements with the objective in an IRL method, both of which are jointly optimized via gradient descent. In the experiments, we show our framework performs safer compared to IRL methods without CBF, that is $\sim15\%$ and $\sim20\%$ improvement for two levels of difficulty of a 2D racecar domain and $\sim 50\%$ improvement for a 3D drone domain.
translated by 谷歌翻译
强化学习(RL)已在域中展示有效,在域名可以通过与其操作环境进行积极互动来学习政策。但是,如果我们将RL方案更改为脱机设置,代理商只能通过静态数据集更新其策略,其中脱机强化学习中的一个主要问题出现,即分配转移。我们提出了一种悲观的离线强化学习(PESSORL)算法,以主动引导代理通过操纵价值函数来恢复熟悉的区域。我们专注于由分销外(OOD)状态引起的问题,并且故意惩罚训练数据集中不存在的状态的高值,以便学习的悲观值函数下限界限状态空间内的任何位置。我们在各种基准任务中评估Pessorl算法,在那里我们表明我们的方法通过明确处理OOD状态,与这些方法仅考虑ood行动时,我们的方法通过明确处理OOD状态。
translated by 谷歌翻译
Imitation learning techniques aim to mimic human behavior in a given task. An agent (a learning machine) is trained to perform a task from demonstrations by learning a mapping between observations and actions. The idea of teaching by imitation has been around for many years, however, the field is gaining attention recently due to advances in computing and sensing as well as rising demand for intelligent applications. The paradigm of learning by imitation is gaining popularity because it facilitates teaching complex tasks with minimal expert knowledge of the tasks. Generic imitation learning methods could potentially reduce the problem of teaching a task to that of providing demonstrations; without the need for explicit programming or designing reward functions specific to the task. Modern sensors are able to collect and transmit high volumes of data rapidly, and processors with high computational power allow fast processing that maps the sensory data to actions in a timely manner. This opens the door for many potential AI applications that require real-time perception and reaction such as humanoid robots, self-driving vehicles, human computer interaction and computer games to name a few. However, specialized algorithms are needed to effectively and robustly learn models as learning by imitation poses its own set of challenges. In this paper, we survey imitation learning methods and present design options in different steps of the learning process. We introduce a background and motivation for the field as well as highlight challenges specific to the imitation problem. Methods for designing and evaluating imitation learning tasks are categorized and reviewed. Special attention is given to learning methods in robotics and games as these domains are the most popular in the literature and provide a wide array of problems and methodologies. We extensively discuss combining imitation learning approaches using different sources and methods, as well as incorporating other motion learning methods to enhance imitation. We also discuss the potential impact on industry, present major applications and highlight current and future research directions.
translated by 谷歌翻译
我们考虑创建助手的问题,这些助手可以帮助代理人(通常是人类)解决新颖的顺序决策问题,假设代理人无法将奖励功能明确指定给助手。我们没有像目前的方法那样旨在自动化并代替代理人,而是赋予助手一个咨询角色,并将代理商作为主要决策者。困难是,我们必须考虑由代理商的限制或限制引起的潜在偏见,这可能导致其看似非理性地拒绝建议。为此,我们介绍了一种新颖的援助形式化,以模拟这些偏见,从而使助手推断和适应它们。然后,我们引入了一种计划助手建议的新方法,该方法可以扩展到大型决策问题。最后,我们通过实验表明我们的方法适应了这些代理偏见,并且比基于自动化的替代方案给代理带来了更高的累积奖励。
translated by 谷歌翻译
马尔可夫决策过程(MDP)是在顺序决策中常用的正式模型。 MDP捕获了可能出现的随机性,例如,通过过渡函数中的概率从不精确的执行器中捕获。但是,在数据驱动的应用程序中,从(有限)数据中得出精确的概率引入了可能导致意外或不良结果的统计错误。不确定的MDP(UMDP)不需要精确的概率,而是在过渡中使用所谓的不确定性集,占此类有限的数据。正式验证社区中的工具有效地计算了强大的政策,这些政策在不确定性集中最坏的情况下,可以证明遵守正式规格,例如安全限制。我们不断地以强大的学习方法与将专用的贝叶斯推理方案与强大策略的计算结合在一起的任何时间学习方法中不断学习MDP的过渡概率。特别是,我们的方法(1)将概率近似为间隔,(2)适应可能与中间模型不一致的新数据,并且可以随时停止(3),以在UMDP上计算强大的策略,以忠实地捕获稳健的策略到目前为止的数据。我们展示了我们的方法的有效性,并将其与在几个基准的实验评估中对UMDP计算出的UMDP进行了比较。
translated by 谷歌翻译