汤普森采样是上下文匪徒的最有效方法之一,已被推广到某些MDP设置后的后验采样。但是,现有的后验学习方法是基于模型或缺乏线性MDP以外的最坏情况的理论保证而受到限制的。本文提出了一种新的无模型后取样公式,该公式适用于具有理论保证的更通用的情节增强学习问题。我们介绍了新颖的证明技术,以表明在适当的条件下,我们的后抽样方法的最遗憾与基于优化的方法的最著名结果相匹配。在具有尺寸的线性MDP设置中,与现有基于后采样的探索算法的二次依赖性相比,我们算法的遗憾与维度线性缩放。
translated by 谷歌翻译
We study sample efficient reinforcement learning (RL) under the general framework of interactive decision making, which includes Markov decision process (MDP), partially observable Markov decision process (POMDP), and predictive state representation (PSR) as special cases. Toward finding the minimum assumption that empowers sample efficient learning, we propose a novel complexity measure, generalized eluder coefficient (GEC), which characterizes the fundamental tradeoff between exploration and exploitation in online interactive decision making. In specific, GEC captures the hardness of exploration by comparing the error of predicting the performance of the updated policy with the in-sample training error evaluated on the historical data. We show that RL problems with low GEC form a remarkably rich class, which subsumes low Bellman eluder dimension problems, bilinear class, low witness rank problems, PO-bilinear class, and generalized regular PSR, where generalized regular PSR, a new tractable PSR class identified by us, includes nearly all known tractable POMDPs. Furthermore, in terms of algorithm design, we propose a generic posterior sampling algorithm, which can be implemented in both model-free and model-based fashion, under both fully observable and partially observable settings. The proposed algorithm modifies the standard posterior sampling algorithm in two aspects: (i) we use an optimistic prior distribution that biases towards hypotheses with higher values and (ii) a loglikelihood function is set to be the empirical loss evaluated on the historical data, where the choice of loss function supports both model-free and model-based learning. We prove that the proposed algorithm is sample efficient by establishing a sublinear regret upper bound in terms of GEC. In summary, we provide a new and unified understanding of both fully observable and partially observable RL.
translated by 谷歌翻译
Epsilon-Greedy,SoftMax或Gaussian噪声等近视探索政策在某些强化学习任务中无法有效探索,但是在许多其他方面,它们的表现都很好。实际上,实际上,由于简单性,它们通常被选为最佳选择。但是,对于哪些任务执行此类政策成功?我们可以为他们的有利表现提供理论保证吗?尽管这些政策具有显着的实际重要性,但这些关键问题几乎没有得到研究。本文介绍了对此类政策的理论分析,并为通过近视探索提供了对增强学习的首次遗憾和样本复杂性。我们的结果适用于具有有限的Bellman Eluder维度的情节MDP中的基于价值功能的算法。我们提出了一种新的复杂度度量,称为近视探索差距,用Alpha表示,该差距捕获了MDP的结构属性,勘探策略和给定的值函数类别。我们表明,近视探索的样品复杂性与该数量的倒数1 / alpha^2二次地量表。我们通过具体的例子进一步证明,由于相应的动态和奖励结构,在近视探索成功的几项任务中,近视探索差距确实是有利的。
translated by 谷歌翻译
我们提出了一个通用框架,以设计基于模型的RL的后验采样方法。我们表明,可以通过减少基于Hellinger距离的条件概率估计的遗憾来分析所提出的算法。我们进一步表明,当我们通过数据可能性测量模型误差时,乐观的后采样可以控制此Hellinger距离。该技术使我们能够设计和分析许多基于模型的RL设置的最先进的样品复杂性保证的统一后采样算法。我们在许多特殊情况下说明了我们的总体结果,证明了我们框架的多功能性。
translated by 谷歌翻译
Despite the significant interest and progress in reinforcement learning (RL) problems with adversarial corruption, current works are either confined to the linear setting or lead to an undesired $\tilde{O}(\sqrt{T}\zeta)$ regret bound, where $T$ is the number of rounds and $\zeta$ is the total amount of corruption. In this paper, we consider the contextual bandit with general function approximation and propose a computationally efficient algorithm to achieve a regret of $\tilde{O}(\sqrt{T}+\zeta)$. The proposed algorithm relies on the recently developed uncertainty-weighted least-squares regression from linear contextual bandit \citep{he2022nearly} and a new weighted estimator of uncertainty for the general function class. In contrast to the existing analysis that heavily relies on the linear structure, we develop a novel technique to control the sum of weighted uncertainty, thus establishing the final regret bounds. We then generalize our algorithm to the episodic MDP setting and first achieve an additive dependence on the corruption level $\zeta$ in the scenario of general function approximation. Notably, our algorithms achieve regret bounds either nearly match the performance lower bound or improve the existing methods for all the corruption levels and in both known and unknown $\zeta$ cases.
translated by 谷歌翻译
We study time-inhomogeneous episodic reinforcement learning (RL) under general function approximation and sparse rewards. We design a new algorithm, Variance-weighted Optimistic $Q$-Learning (VO$Q$L), based on $Q$-learning and bound its regret assuming completeness and bounded Eluder dimension for the regression function class. As a special case, VO$Q$L achieves $\tilde{O}(d\sqrt{HT}+d^6H^{5})$ regret over $T$ episodes for a horizon $H$ MDP under ($d$-dimensional) linear function approximation, which is asymptotically optimal. Our algorithm incorporates weighted regression-based upper and lower bounds on the optimal value function to obtain this improved regret. The algorithm is computationally efficient given a regression oracle over the function class, making this the first computationally tractable and statistically optimal approach for linear MDPs.
translated by 谷歌翻译
Modern Reinforcement Learning (RL) is commonly applied to practical problems with an enormous number of states, where function approximation must be deployed to approximate either the value function or the policy. The introduction of function approximation raises a fundamental set of challenges involving computational and statistical efficiency, especially given the need to manage the exploration/exploitation tradeoff. As a result, a core RL question remains open: how can we design provably efficient RL algorithms that incorporate function approximation? This question persists even in a basic setting with linear dynamics and linear rewards, for which only linear function approximation is needed.This paper presents the first provable RL algorithm with both polynomial runtime and polynomial sample complexity in this linear setting, without requiring a "simulator" or additional assumptions. Concretely, we prove that an optimistic modification of Least-Squares Value Iteration (LSVI)-a classical algorithm frequently studied in the linear setting-achieves O( √ d 3 H 3 T ) regret, where d is the ambient dimension of feature space, H is the length of each episode, and T is the total number of steps. Importantly, such regret is independent of the number of states and actions.
translated by 谷歌翻译
信息指导的采样(IDS)揭示了其作为增强学习(RL)的数据效率算法的潜力。但是,对马尔可夫决策过程(MDP)的ID的理论理解仍然有限。我们开发了新颖的信息理论工具,以限制有关学习目标的信息比和累积信息获得。我们的理论结果阐明了选择学习目标的重要性,以便从业者可以平衡计算和后悔的界限。结果,我们为香草IDS提供了先前的贝叶斯遗憾界限,该范围在表格有限的摩尼子MDP下学习了整个环境。此外,我们提出了一种计算效率的正规化ID,该ID可以最大化添加剂形式而不是比率形式,并表明它具有与香草-IDS相同的遗憾。借助利率延伸理论,我们通过学习一个代孕,信息不足的环境来改善遗憾。此外,我们将分析扩展到线性MDP,并证明了汤普森采样作为副产品的类似遗憾界限。
translated by 谷歌翻译
寻找统一的复杂性度量和样本效率学习的算法是增强学习研究的核心主题(RL)。 Foster等人最近提出了决策估计系数(DEC)。 (2021)作为样品有效的NO-REGRET RL的必要和足够的复杂度度量。本文通过DEC框架朝着RL的统一理论取得了进步。首先,我们提出了两项​​新的DEC类型复杂性度量:探索性DEC(EDEC)和无奖励DEC(RFDEC)。我们表明,它们对于样本有效的PAC学习和无奖励学习是必要的,因此扩展了原始DEC,该DEC仅捕获了无需重新学习。接下来,我们为所有三个学习目标设计新的统一样品效率算法。我们的算法实例化估计到决策的变体(E2D)元算法具有强大而通用的模型估计值。即使在无重组的设置中,我们的算法E2D-TA也会在Foster等人的算法上提高。 (2021)需要对DEC的变体进行边界,该变体可能是过于大的,或者设计特定问题的估计值。作为应用程序,我们恢复了现有的,并获得了使用单个算法的各种可拖动RL问题的新样品学习结果。最后,作为一种连接,我们根据后采样或最大似然估计重新分析了两种现有的基于乐观模型的算法,表明它们在与DEC相似的结构条件下具有与E2D-TA相似的遗憾界限。
translated by 谷歌翻译
我们在非静止线性(AKA低级别)马尔可夫决策过程(MDP)中研究了集中加强学习,即奖励和转换内核都是关于给定特征映射的线性,并且被允许缓慢或突然演变时间。对于此问题设置,我们提出了一种基于加权最小二乘值的乐观模型算法的Opt-WLSVI,其使用指数权重来平滑地忘记过去远远的数据。我们表明我们的算法在每次竞争最佳政策时,实现了由$ \ widetilde {\ mathcal {o}}的上部界限的遗憾(d ^ {5/4} h ^ 2 \ delta ^ {1 / 4} k ^ {3/4})$何地在$ d $是特征空间的尺寸,$ h $是规划地平线,$ k $是剧集的数量和$ \ delta $是一个合适的衡量标准MDP的非固定性。此外,我们指出了在忘记以前作品的非静止线性匪徒环境中忘记策略的技术差距,并提出了修复其遗憾分析。
translated by 谷歌翻译
我们研究了受限的强化学习问题,其中代理的目的是最大程度地提高预期的累积奖励,从而受到对实用程序函数的预期总价值的约束。与现有的基于模型的方法或无模型方法伴随着“模拟器”,我们旨在开发第一个无模型的无模拟算法,即使在大规模系统中,也能够实现sublinear遗憾和透明度的约束侵犯。为此,我们考虑具有线性函数近似的情节约束决策过程,其中过渡动力学和奖励函数可以表示为某些已知功能映射的线性函数。我们表明$ \ tilde {\ mathcal {o}}(\ sqrt {d^3h^3t})$遗憾和$ \ tilde {\ tillcal {\ mathcal {o}}(\ sqrt {d^3h^3ht})$约束$约束$约束可以实现违规范围,其中$ d $是功能映射的尺寸,$ h $是情节的长度,而$ t $是总数的总数。我们的界限是在没有明确估计未知过渡模型或需要模拟器的情况下达到的,并且仅通过特征映射的维度依赖于状态空间。因此,即使国家的数量进入无穷大,我们的界限也会存在。我们的主要结果是通过标准LSVI-UCB算法的新型适应来实现的。特别是,我们首先将原始二次优化引入LSVI-UCB算法中,以在遗憾和违反约束之间取得平衡。更重要的是,我们使用软马克斯政策取代了LSVI-UCB中的状态行动功能的标准贪婪选择。事实证明,这对于通过其近似平滑度的权衡来确定受约束案例的统一浓度是关键。我们还表明,一个人可以达到均匀的约束违规行为,同时仍然保持相同的订单相对于$ t $。
translated by 谷歌翻译
我们使用访问离线最小二乘回归甲骨文的访问权限,在最低可及性假设下为随机上下文MDP提供了遗憾的最小化算法。我们分析了三个不同的设置:在该动力学的位置,动力学是未知的,但独立于上下文和最具挑战性的设置,而动力学是未知和上下文依赖性的。对于后者,我们的算法获得$ \ tilde {o} \ left(\ max \ {h,{1}/{p_ {min}}} \} \} t \ log(\ max \ {| \ mathcal {f} |,| \ mathcal {p} | \}/\ delta)} \ right)$ hearse bunder bund bund bund bund bund bund bund bunging bund bunger,probinality $ 1- \ delta $,其中$ \ mathcal { P} $和$ \ Mathcal {f} $是用于分别近似动态和奖励的有限且可实现的函数类,$ p_ {min} $是最小可及性参数,$ s $是一组状态,$ a $ a $一组动作,$ h $ the Horizo​​n和$ t $情节数。据我们所知,我们的方法是使用一般函数近似的上下文MDP的第一种乐观方法(即,在没有其他有关功能类别的知识的情况下,例如线性等)。此外,我们还提供$ \ omega的下限即使在已知的动态情况下,也会产生预期的遗憾。
translated by 谷歌翻译
在线学习和决策中的一个核心问题 - 从土匪到强化学习 - 是要了解哪种建模假设会导致样本有效的学习保证。我们考虑了一个普遍的对抗性决策框架,该框架涵盖了(结构化的)匪徒问题,这些问题与对抗性动力学有关。我们的主要结果是通过新的上限和下限显示决策估计系数,这是Foster等人引入的复杂度度量。在与我们环境的随机对应物中,对于对抗性决策而言是必要和足够的遗憾。但是,与随机设置相比,必须将决策估计系数应用于所考虑的模型类(或假设)的凸壳。这就确定了容纳对抗奖励或动态的价格受凸层化模型类的行为的约束,并恢复了许多现有结果 - 既积极又负面。在获得这些保证的途径中,我们提供了新的结构结果,将决策估计系数与其他众所周知的复杂性度量的变体联系起来,包括Russo和Van Roy的信息比以及Lattimore和Gy的探索目标\“ {o} rgy。
translated by 谷歌翻译
本文提供了一项理论研究,该研究对在线环境下的$ \ epsilon $ - 梅迪探索中的增强学习(RL)中的深神经功能近似(RL)提供了研究。这种问题设置是由属于该制度的成功深Q-Networks(DQN)框架所激发的。在这项工作中,我们从函数类别和神经网络体系结构(例如,宽度和深度)的角度从“线性”制度之外的函数类别和神经网络体系结构(例如宽度和深度)提供了对理论理解的初步尝试。具体来说,我们将重点放在基于价值的算法上,分别通过BESOV(和Barron)功能空间赋予的深层(和两层)神经网络,以$ \ epsilon $ greedy探索,旨在近似于$ \ alpha $ -Smooth Q功能在$ d $二维功能空间中。我们证明,使用$ t $情节,缩放宽度$ m = \ widetilde {\ mathcal {o}}}(t^{\ frac {d} {2 \ alpha + d}})$和depth $ l = \ Mathcal {O}(\ log t)for Deep RL的神经网络的$足以在Besov空间中以sublinear的遗憾学习。此外,对于由Barron空间赋予的两层神经网络,缩放宽度$ \ omega(\ sqrt {t})$就足够了。为了实现这一目标,我们分析中的关键问题是如何估计深神经功能近似下的时间差异误差,因为$ \ epsilon $ - 否则探索不足以确保“乐观”。我们的分析重新制定了$ l^2(\ mathrm {d} \ mu)$ - 在某个平均度量$ \ mu $上的可集成空间,并将其转换为非IID设置下的概括问题。这可能对RL理论具有自身的兴趣,以便更好地理解Deep RL中的$ \ Epsilon $ -Greedy Exploration。
translated by 谷歌翻译
代表学习呈现在深入学习的经验成功的核心,以处理维度的诅咒。然而,由于i),表现力(RL)的钢筋学习(RL)尚未充分利用卓越的能力,表现力和易疏忽之间的权衡;二世),探索与代表学习之间的耦合。在本文中,我们首先揭示了在随机控制模型中的一些噪声假设下,我们可以免费获得其相应的马尔可夫过渡操作员的线性谱特征。基于该观察,我们提出了嵌入(Spede)的谱动力学嵌入(SPEDE),这将通过利用噪声结构来完成对代表学习的乐观探索。我们提供对Speded的严格理论分析,并展示了几种基准上现有最先进的实证算法的实际卓越性能。
translated by 谷歌翻译
政策优化方法是使用最广泛使用的加固学习(RL)算法之一。然而,对这些方法的理论理解仍然不足。即使在eoisodic(时代)的表格设置中,\ citet的基于政策方法的最先进的理论结果也是只需$ \ tilde {o}(\ sqrt {s ^ 2ah ^ 4k })$何地在$ S $是州的数量,$ a $是行动的数量,$ h $是地平线,而$ k $是剧集的数量,还有$ \ sqrt {sh} $与信息理论下限$ \ tilde {\ omega}相比,差距(\ sqrt {sah ^ 3k})$。为了弥合这样的差距,我们提出了一种新的算法基于参考的基于参考的策略优化,在任何时间保证(\ AlgnameAcro),它具有“随时稳定”的特征。我们证明我们的算法实现$ \ tilde {o}(\ sqrt {sah ^ 3k} + \ sqrt {ah ^ 4})$后悔。当$ s> h $时,我们的算法在忽略对数因子时最佳最佳。为了我们的最佳知识,RPO-SAT是第一次计算上高效,几乎最小的表格RL最佳策略算法。
translated by 谷歌翻译
在许多综合设置(例如视频游戏)和GO中,增强学习(RL)超出了人类的绩效。但是,端到端RL模型的现实部署不太常见,因为RL模型对环境的轻微扰动非常敏感。强大的马尔可夫决策过程(MDP)框架(其中的过渡概率属于名义模型设置的不确定性)提供了一种开发健壮模型的方法。虽然先前的分析表明,RL算法是有效的,假设访问生成模型,但尚不清楚RL在更现实的在线设置下是否可以有效,这需要在探索和开发之间取得仔细的平衡。在这项工作中,我们通过与未知的名义系统进行互动来考虑在线强大的MDP。我们提出了一种强大的乐观策略优化算法,该算法可有效。为了解决由对抗性环境引起的其他不确定性,我们的模型具有通过Fenchel Conjugates得出的新的乐观更新规则。我们的分析确定了在线强大MDP的第一个遗憾。
translated by 谷歌翻译
Value-function approximation methods that operate in batch mode have foundational importance to reinforcement learning (RL). Finite sample guarantees for these methods often crucially rely on two types of assumptions: (1) mild distribution shift, and (2) representation conditions that are stronger than realizability. However, the necessity ("why do we need them?") and the naturalness ("when do they hold?") of such assumptions have largely eluded the literature. In this paper, we revisit these assumptions and provide theoretical results towards answering the above questions, and make steps towards a deeper understanding of value-function approximation.
translated by 谷歌翻译
Model-free reinforcement learning (RL) algorithms, such as Q-learning, directly parameterize and update value functions or policies without explicitly modeling the environment. They are typically simpler, more flexible to use, and thus more prevalent in modern deep RL than model-based approaches. However, empirical work has suggested that model-free algorithms may require more samples to learn [7,22]. The theoretical question of "whether model-free algorithms can be made sample efficient" is one of the most fundamental questions in RL, and remains unsolved even in the basic scenario with finitely many states and actions.We prove that, in an episodic MDP setting, Q-learning with UCB exploration achieves regret Õ( √ H 3 SAT ), where S and A are the numbers of states and actions, H is the number of steps per episode, and T is the total number of steps. This sample efficiency matches the optimal regret that can be achieved by any model-based approach, up to a single √ H factor. To the best of our knowledge, this is the first analysis in the model-free setting that establishes √ T regret without requiring access to a "simulator." * The first two authors contributed equally.
translated by 谷歌翻译
We develop an extension of posterior sampling for reinforcement learning (PSRL) that is suited for a continuing agent-environment interface and integrates naturally into agent designs that scale to complex environments. The approach maintains a statistically plausible model of the environment and follows a policy that maximizes expected $\gamma$-discounted return in that model. At each time, with probability $1-\gamma$, the model is replaced by a sample from the posterior distribution over environments. For a suitable schedule of $\gamma$, we establish an $\tilde{O}(\tau S \sqrt{A T})$ bound on the Bayesian regret, where $S$ is the number of environment states, $A$ is the number of actions, and $\tau$ denotes the reward averaging time, which is a bound on the duration required to accurately estimate the average reward of any policy.
translated by 谷歌翻译