Real-world reinforcement learning tasks often involve some form of partial observability where the observations only give a partial or noisy view of the true state of the world. Such tasks typically require some form of memory, where the agent has access to multiple past observations, in order to perform well. One popular way to incorporate memory is by using a recurrent neural network to access the agent's history. However, recurrent neural networks in reinforcement learning are often fragile and difficult to train, susceptible to catastrophic forgetting and sometimes fail completely as a result. In this work, we propose Deep Transformer Q-Networks (DTQN), a novel architecture utilizing transformers and self-attention to encode an agent's history. DTQN is designed modularly, and we compare results against several modifications to our base model. Our experiments demonstrate the transformer can solve partially observable tasks faster and more stably than previous recurrent approaches.

Transformer, originally devised for natural language processing, has also attested significant success in computer vision. Thanks to its super expressive power, researchers are investigating ways to deploy transformers to reinforcement learning (RL) and the transformer-based models have manifested their potential in representative RL benchmarks. In this paper, we collect and dissect recent advances on transforming RL by transformer (transformer-based RL or TRL), in order to explore its development trajectory and future trend. We group existing developments in two categories: architecture enhancement and trajectory optimization, and examine the main applications of TRL in robotic manipulation, text-based games, navigation and autonomous driving. For architecture enhancement, these methods consider how to apply the powerful transformer structure to RL problems under the traditional RL framework, which model agents and environments much more precisely than deep RL methods, but they are still limited by the inherent defects of traditional RL algorithms, such as bootstrapping and "deadly triad". For trajectory optimization, these methods treat RL problems as sequence modeling and train a joint state-action model over entire trajectories under the behavior cloning framework, which are able to extract policies from static datasets and fully use the long-sequence modeling capability of the transformer. Given these advancements, extensions and challenges in TRL are reviewed and proposals about future direction are discussed. We hope that this survey can provide a detailed introduction to TRL and motivate future research in this rapidly developing field.

通过回顾一封来自情节记忆的过去的经验，可以通过回忆过去的经验来实现钢筋学习的样本效率。我们提出了一种新的基于模型的轨迹的集体记忆，解决了集体控制的当前限制。我们的记忆估计轨迹值，指导代理人朝着良好的政策。基于内存构建，我们通过动态混合控制统一模型的基于动态和习惯学习来构建互补学习模型，进入单个架构。实验表明，我们的模型可以比各种环境中的其他强力加强学习代理更快，更好地学习，包括随机和非马尔可夫环境。

在部分可观察到的马尔可夫决策过程（POMDP）中，代理通常使用过去的表示来近似基础MDP。我们建议利用冷冻验证的语言变压器（PLT）进行病史表示和压缩，以提高样品效率。为了避免对变压器进行训练，我们引入了Frozenhopfield，该菲尔德自动将观察结果与预处理的令牌嵌入相关联。为了形成这些关联，现代的Hopfield网络存储了这些令牌嵌入，这些嵌入是通过查询获得的查询来检索的，这些嵌入者通过随机但固定的观察结果获得。我们的新方法Helm，启用了Actor-Critic网络体系结构，该架构包含用于历史记录表示的历史模块的审计语言变压器。由于不需要学习过去的代表，因此掌舵比竞争对手要高得多。在Miligrid和Procgen环境上，Helm掌舵取得了新的最新结果。我们的代码可在https://github.com/ml-jku/helm上找到。

微调加强学习（RL）模型由于缺乏大规模的现成数据集以及不同环境之间可传递性的较高差异而变得具有挑战性。最近的工作着眼于从序列建模的角度来应对离线RL，并通过引入变压器体系结构的结果得到改进的结果。但是，当模型从头开始训练时，它会遭受缓慢的收敛速度。在本文中，我们希望利用这种强化学习作为序列建模的表述，并研究在离线RL任务（控制，游戏）上进行填充时，在其他领域（视觉，语言）上进行了预训练的序列模型的可传递性。为此，我们还提出了改善这些域之间传递的技术。结果表明，在各种环境上的收敛速度和奖励方面，表现出一致的性能，加速了3-6倍的训练，并使用Wikipedia-pretrenained and GPT2语言模型在各种任务中实现了最先进的绩效。我们希望这项工作不仅为RL利用通用序列建模技术和预训练模型的潜力带来启发，而且还激发了未来的工作，在完全不同领域的生成建模任务之间共享知识。

Deep reinforcement learning is poised to revolutionise the field of AI and represents a step towards building autonomous systems with a higher level understanding of the visual world. Currently, deep learning is enabling reinforcement learning to scale to problems that were previously intractable, such as learning to play video games directly from pixels. Deep reinforcement learning algorithms are also applied to robotics, allowing control policies for robots to be learned directly from camera inputs in the real world. In this survey, we begin with an introduction to the general field of reinforcement learning, then progress to the main streams of value-based and policybased methods. Our survey will cover central algorithms in deep reinforcement learning, including the deep Q-network, trust region policy optimisation, and asynchronous advantage actor-critic. In parallel, we highlight the unique advantages of deep neural networks, focusing on visual understanding via reinforcement learning. To conclude, we describe several current areas of research within the field.

为了解决艰巨的任务，人类提出问题以从外部来源获取知识。相反，经典的加强学习者缺乏这种能力，并且常常诉诸探索性行为。这会加剧，因为很少的当今环境支持查询知识。为了研究如何通过语言教授代理来查询外部知识，我们首先介绍了两个新环境：基于网格世界的Q-babyai和基于文本的Q-Textworld。除了物理互动外，代理还可以查询专门针对这些环境的外部知识源来收集信息。其次，我们提出了“寻求知识”（AFK）代理，该代理学会生成语言命令以查询有助于解决任务的有意义的知识。 AFK利用非参数记忆，指针机制和情节探索奖金来解决（1）无关的信息，（2）一个较大的查询语言空间，（3）延迟奖励有意义的查询。广泛的实验表明，AFK代理在具有挑战性的Q-Babyai和Q-Textworld环境方面优于最近的基线。

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。

最先进的多机构增强学习（MARL）方法为各种复杂问题提供了有希望的解决方案。然而，这些方法都假定代理执行同步的原始操作执行，因此它们不能真正可扩展到长期胜利的真实世界多代理/机器人任务，这些任务固有地要求代理/机器人以异步的理由，涉及有关高级动作选择的理由。不同的时间。宏观行动分散的部分可观察到的马尔可夫决策过程（MACDEC-POMDP）是在完全合作的多代理任务中不确定的异步决策的一般形式化。在本论文中，我们首先提出了MacDec-Pomdps的一组基于价值的RL方法，其中允许代理在三个范式中使用宏观成果功能执行异步学习和决策：分散学习和控制，集中学习，集中学习和控制，以及分散执行的集中培训（CTDE）。在上述工作的基础上，我们在三个训练范式下制定了一组基于宏观行动的策略梯度算法，在该训练范式下，允许代理以异步方式直接优化其参数化策略。我们在模拟和真实的机器人中评估了我们的方法。经验结果证明了我们在大型多代理问题中的方法的优势，并验证了我们算法在学习具有宏观actions的高质量和异步溶液方面的有效性。

我们提出了一个端到端，基于模型的深度加强学习代理，它在规划期间动态地参加其国家的相关部分。代理使用基于集的表示的瓶颈机制，以强制代理参加每个规划步骤的实体数量。在实验中，我们研究了具有不同挑战的几套定制环境的瓶颈机制。我们始终如一地观察到该设计允许规划代理通过参加相关对象来概括其在兼容的看不见环境中的学习任务解决能力，从而导致更好的分发概括性表现。

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.

Text-based games present a unique class of sequential decision making problem in which agents interact with a partially observable, simulated environment via actions and observations conveyed through natural language. Such observations typically include instructions that, in a reinforcement learning (RL) setting, can directly or indirectly guide a player towards completing reward-worthy tasks. In this work, we study the ability of RL agents to follow such instructions. We conduct experiments that show that the performance of state-of-the-art text-based game agents is largely unaffected by the presence or absence of such instructions, and that these agents are typically unable to execute tasks to completion. To further study and address the task of instruction following, we equip RL agents with an internal structured representation of natural language instructions in the form of Linear Temporal Logic (LTL), a formal language that is increasingly used for temporally extended reward specification in RL. Our framework both supports and highlights the benefit of understanding the temporal semantics of instructions and in measuring progress towards achievement of such a temporally extended behaviour. Experiments with 500+ games in TextWorld demonstrate the superior performance of our approach.

Inspired by progress in large-scale language modeling, we apply a similar approach towards building a single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based on its context whether to output text, joint torques, button presses, or other tokens. In this report we describe the model and the data, and document the current capabilities of Gato.

We present CURL: Contrastive Unsupervised Representations for Reinforcement Learning. CURL extracts high-level features from raw pixels using contrastive learning and performs offpolicy control on top of the extracted features. CURL outperforms prior pixel-based methods, both model-based and model-free, on complex tasks in the DeepMind Control Suite and Atari Games showing 1.9x and 1.2x performance gains at the 100K environment and interaction steps benchmarks respectively. On the DeepMind Control Suite, CURL is the first image-based algorithm to nearly match the sample-efficiency of methods that use state-based features. Our code is open-sourced and available at https://www. github.com/MishaLaskin/curl.

Designing better deep networks and better reinforcement learning (RL) algorithms are both important for deep RL. This work focuses on the former. Previous methods build the network with several modules like CNN, LSTM and Attention. Recent methods combine the Transformer with these modules for better performance. However, it requires tedious optimization skills to train a network composed of mixed modules, making these methods inconvenient to be used in practice. In this paper, we propose to design \emph{pure Transformer-based networks} for deep RL, aiming at providing off-the-shelf backbones for both the online and offline settings. Specifically, the Transformer in Transformer (TIT) backbone is proposed, which cascades two Transformers in a very natural way: the inner one is used to process a single observation, while the outer one is responsible for processing the observation history; combining both is expected to extract spatial-temporal representations for good decision-making. Experiments show that TIT can achieve satisfactory performance in different settings, consistently.

We propose a conceptually simple and lightweight framework for deep reinforcement learning that uses asynchronous gradient descent for optimization of deep neural network controllers. We present asynchronous variants of four standard reinforcement learning algorithms and show that parallel actor-learners have a stabilizing effect on training allowing all four methods to successfully train neural network controllers. The best performing method, an asynchronous variant of actor-critic, surpasses the current state-of-the-art on the Atari domain while training for half the time on a single multi-core CPU instead of a GPU. Furthermore, we show that asynchronous actor-critic succeeds on a wide variety of continuous motor control problems as well as on a new task of navigating random 3D mazes using a visual input.

基于文本的游戏（TBG）是复杂的环境，允许用户或计算机代理进行文本交互并实现游戏目标。为基于文本的游戏构建面向目标的计算机代理是一项挑战，尤其是当我们使用逐步反馈作为模型的唯一文本输入时。此外，代理商很难通过从更大的文本输入空间中评估灵活的长度和形式。在本文中，我们对应用于基于文本的游戏字段的深度学习方法进行了广泛的分析。

Starcraft II多代理挑战（SMAC）被创建为合作多代理增强学习（MARL）的具有挑战性的基准问题。 SMAC专注于星际争霸微管理的问题，并假设每个单元都由独立行动并仅具有本地信息的学习代理人单独控制；假定通过分散执行（CTDE）进行集中培训。为了在SMAC中表现良好，MARL算法必须处理多机构信贷分配和联合行动评估的双重问题。本文介绍了一种新的体系结构Transmix，这是一个基于变压器的联合行动值混合网络，与其他最先进的合作MARL解决方案相比，我们显示出高效且可扩展的。 Transmix利用变形金刚学习更丰富的混合功能的能力来结合代理的个人价值函数。它与以前的SMAC场景上的工作相当，并且在困难场景上胜过其他技术，以及被高斯噪音损坏的场景以模拟战争的雾。

当相互作用数据稀缺时，深厚的增强学习（RL）算法遭受了严重的性能下降，这限制了其现实世界的应用。最近，视觉表示学习已被证明是有效的，并且有望提高RL样品效率。这些方法通常依靠对比度学习和数据扩展来训练状态预测的过渡模型，这与在RL中使用模型的方式不同 - 基于价值的计划。因此，学到的模型可能无法与环境保持良好状态并产生一致的价值预测，尤其是当国家过渡不是确定性的情况下。为了解决这个问题，我们提出了一种称为价值一致表示学习（VCR）的新颖方法，以学习与决策直接相关的表示形式。更具体地说，VCR训练一个模型，以预测基于当前的状态（也称为“想象的状态”）和一系列动作。 VCR没有将这个想象中的状态与环境返回的真实状态保持一致，而是在两个状态上应用$ q $ - 价值头，并获得了两个行动值分布。然后将距离计算并最小化以迫使想象的状态产生与真实状态相似的动作值预测。我们为离散和连续的动作空间开发了上述想法的两个实现。我们对Atari 100K和DeepMind Control Suite基准测试进行实验，以验证其提高样品效率的有效性。已经证明，我们的方法实现了无搜索RL算法的新最新性能。