Experience replay is widely used in deep reinforcement learning algorithms and allows agents to remember and learn from experiences from the past. In an effort to learn more efficiently, researchers proposed prioritized experience replay (PER) which samples important transitions more frequently. In this paper, we propose Prioritized Sequence Experience Replay (PSER) a framework for prioritizing sequences of experience in an attempt to both learn more efficiently and to obtain better performance. We compare the performance of PER and PSER sampling techniques in a tabular Q-learning environment and in DQN on the Atari 2600 benchmark. We prove theoretically that PSER is guaranteed to converge faster than PER and empirically show PSER substantially improves upon PER.
Korea Advanced Institute of Science and Technology（韩国科学技术院）
Experience replay enables online reinforcement learning agents to store and reuse the previous experiences of interacting with the environment. In the original method, the experiences are sampled and replayed uniformly at random. A prior work called prioritized experience replay was developed where experiences are prioritized, so as to replay experiences seeming to be more important more frequently. In this paper, we develop a method called double-prioritized state-recycled (DPSR) experience replay, prioritizing the experiences in both training stage and storing stage, as well as replacing the…
Experience replay lets online reinforcement learning agents remember and reuse experiences from the past. In prior work, experience transitions were uniformly sampled from a replay memory. However, this approach simply replays transitions at the same frequency that they were originally experienced, regardless of their significance. In this paper we develop a framework for prioritizing experience, so as to replay important transitions more frequently, and therefore learn more efficiently. We use prioritized experience replay in Deep Q-Networks (DQN), a reinforcement learning algorithm that achi…
In goal-conditioned offline reinforcement learning, an agent learns from previously collected data to go to an arbitrary goal. Since the offline data only contains a finite number of trajectories, a main challenge is how to generate more data. Goal-swapping generates additional data by switching trajectory goals but while doing so produces a large number of invalid trajectories. To address this issue, we propose prioritized goal-swapping experience replay (PGSER). PGSER uses a pre-trained Q function to assign higher priority weights to goal swapped transitions that allow reaching the goal. In …
We propose a distributed architecture for deep reinforcement learning at scale, that enables agents to learn effectively from orders of magnitude more data than previously possible. The algorithm decouples acting from learning: the actors interact with their own instances of the environment by selecting actions according to a shared neural network, and accumulate the resulting experience in a shared experience replay memory; the learner replays samples of experience and updates the neural network. The architecture relies on prioritized experience replay to focus only on the most significant da…
Sorbonne Université（索邦大学）；CNRS（法国国家科学研究中心）；Institut des Systèmes Intelligents et de Robotique（智能系统与机器人研究所）
During sleep and awake rest, the hippocampus replays sequences of place cells that have been activated during prior experiences. These have been interpreted as a memory consolidation process, but recent results suggest a possible interpretation in terms of reinforcement learning. The Dyna reinforcement learning algorithms use off-line replays to improve learning. Under limited replay budget, a prioritized sweeping approach, which requires a model of the transitions to the predecessors, can be used to improve performance. We investigate whether such algorithms can explain the experimentally obs…
Prioritized experience replay is a reinforcement learning technique whereby agents speed up learning by replaying useful past experiences. This usefulness is quantified as the expected gain from replaying the experience, a quantity often approximated as the prediction error (TD-error). However, recent work in neuroscience suggests that, in biological organisms, replay is prioritized not only by gain, but also by "need" -- a quantity measuring the expected relevance of each experience with respect to the current situation. Importantly, this term is not currently considered in algorithms such as…
A widely-studied deep reinforcement learning (RL) technique known as Prioritized Experience Replay (PER) allows agents to learn from transitions sampled with non-uniform probability proportional to their temporal-difference (TD) error. Although it has been shown that PER is one of the most crucial components for the overall performance of deep RL methods in discrete action domains, many empirical studies indicate that it considerably underperforms actor-critic algorithms in continuous control. We theoretically show that actor networks cannot be effectively trained with transitions that have la…
Facebook AI Research（脸书人工智能研究院）；University College London（伦敦大学学院）
Environments with procedurally generated content serve as important benchmarks for testing systematic generalization in deep reinforcement learning. In this setting, each level is an algorithmically created environment instance with a unique configuration of its factors of variation. Training on a prespecified subset of levels allows for testing generalization to unseen levels. What can be learned from a level depends on the current policy, yet prior work defaults to uniform sampling of training levels independently of the policy. We introduce Prioritized Level Replay (PLR), a general framewor…
On-policy deep reinforcement learning algorithms have low data utilization and require significant experience for policy improvement. This paper proposes a proximal policy optimization algorithm with prioritized trajectory replay (PTR-PPO) that combines on-policy and off-policy methods to improve sampling efficiency by prioritizing the replay of trajectories generated by old policies. We first design three trajectory priorities based on the characteristics of trajectories: the first two being max and mean trajectory priorities based on one-step empirical generalized advantage estimation (GAE) …
Experience replay enables off-policy reinforcement learning (RL) agents to utilize past experiences to maximize the cumulative reward. Prioritized experience replay that weighs experiences by the magnitude of their temporal-difference error ($|\text{TD}|$) significantly improves the learning efficiency. But how $|\text{TD}|$ is related to the importance of experience is not well understood. We address this problem from an economic perspective, by linking $|\text{TD}|$ to value of experience, which is defined as the value added to the cumulative reward by accessing the experience. We theoretica…
ViZDoom is a robust, first-person shooter reinforcement learning environment, characterized by a significant degree of latent state information. In this paper, double-Q learning and prioritized experience replay methods are tested under a certain ViZDoom combat scenario using a competitive deep recurrent Q-network (DRQN) architecture. In addition, an ensembling technique known as snapshot ensembling is employed using a specific annealed learning rate to observe differences in ensembling efficacy under these two methods. Annealed learning rates are important in general to the training of deep n…
Rollout-Level Advantage-Prioritized Experience Replay for GRPO
Department of Electrical and Computer Engineering, Seoul National University（首尔国立大学电子与计算机工程系）；Interdisciplinary Program in AI, Seoul National University（首尔国立大学人工智能跨学科项目）；AIIS, ASRI, INMC, and ISRC, Seoul National University（首尔国立大学人工智能研究所、人工智能研究机构、智能网络与计算中心及人工智能科学研究中心）
The experience replay mechanism allows agents to use the experiences multiple times. In prior works, the sampling probability of the transitions was adjusted according to their importance. Reassigning sampling probabilities for every transition in the replay buffer after each iteration is highly inefficient. Therefore, experience replay prioritization algorithms recalculate the significance of a transition when the corresponding transition is sampled to gain computational efficiency. However, the importance level of the transitions changes dynamically as the policy and the value function of th…
Bauman Moscow State Technical University（鲍曼莫斯科国立技术大学）
This paper describes an improvement in Deep Q-learning called Reverse Experience Replay (also RER) that solves the problem of sparse rewards and helps to deal with reward maximizing tasks by sampling transitions successively in reverse order. On tasks with enough experience for training and enough Experience Replay memory capacity, Deep Q-learning Network with Reverse Experience Replay shows competitive results against both Double DQN, with a standard Experience Replay, and vanilla DQN. Also, RER achieves significantly increased results in tasks with a lack of experience and Replay memory capa…
We investigate the combination of actor-critic reinforcement learning algorithms with uniform large-scale experience replay and propose solutions for two challenges: (a) efficient actor-critic learning with experience replay (b) stability of off-policy learning where agents learn from other agents behaviour. We employ those insights to accelerate hyper-parameter sweeps in which all participating agents run concurrently and share their experience via a common replay module. To this end we analyze the bias-variance tradeoffs in V-trace, a form of importance sampling for actor-critic methods. Bas…
National Key Laboratory of Novel Software Technology（计算机软件新技术国家重点实验室）；Nanjing University（南京大学）
In reinforcement learning, experience replay stores past samples for further reuse. Prioritized sampling is a promising technique to better utilize these samples. Previous criteria of prioritization include TD error, recentness and corrective feedback, which are mostly heuristically designed. In this work, we start from the regret minimization objective, and obtain an optimal prioritization strategy for Bellman update that can directly maximize the return of the policy. The theory suggests that data with higher hindsight TD error, better on-policiness and more accurate Q value should be assign…
Experience replay enables reinforcement learning agents to memorize and reuse past experiences, just as humans replay memories for the situation at hand. Contemporary off-policy algorithms either replay past experiences uniformly or utilize a rule-based replay strategy, which may be sub-optimal. In this work, we consider learning a replay policy to optimize the cumulative reward. Replay learning is challenging because the replay memory is noisy and large, and the cumulative reward is unstable. To address these issues, we propose a novel experience replay optimization (ERO) framework which alte…
Experience replay is an important technique for addressing sample-inefficiency in deep reinforcement learning (RL), but faces difficulty in learning from binary and sparse rewards due to disproportionately few successful experiences in the replay buffer. Hindsight experience replay (HER) was recently proposed to tackle this difficulty by manipulating unsuccessful transitions, but in doing so, HER introduces a significant bias in the replay buffer experiences and therefore achieves a suboptimal improvement in sample-efficiency. In this paper, we present an analysis on the source of bias in HER,…
Experience replay is a key technique behind many recent advances in deep reinforcement learning. Allowing the agent to learn from earlier memories can speed up learning and break undesirable temporal correlations. Despite its wide-spread application, very little is understood about the properties of experience replay. How does the amount of memory kept affect learning dynamics? Does it help to prioritize certain experiences? In this paper, we address these questions by formulating a dynamical systems ODE model of Q-learning with experience replay. We derive analytic solutions of the ODE for a …
Army Research Laboratory（陆军研究实验室）；Washington State University（华盛顿州立大学）；The University of Texas at Austin（德克萨斯大学奥斯汀分校）；Sony AI（索尼AI）；University of Alberta（阿尔伯塔大学）；Alberta Machine Intelligence Institute（阿尔伯塔机器智能研究所）
Experience replay (ER) improves the data efficiency of off-policy reinforcement learning (RL) algorithms by allowing an agent to store and reuse its past experiences in a replay buffer. While many techniques have been proposed to enhance ER by biasing how experiences are sampled from the buffer, thus far they have not considered strategies for refreshing experiences inside the buffer. In this work, we introduce Lucid Dreaming for Experience Replay (LiDER), a conceptually new framework that allows replay experiences to be refreshed by leveraging the agent's current policy. LiDER consists of thr…
ISAE-SUPAERO（法国高等航空航天学院）；Université de Toulouse（图卢兹大学）；Google Research（谷歌研究院）
Several algorithms have been proposed to sample non-uniformly the replay buffer of deep Reinforcement Learning (RL) agents to speed-up learning, but very few theoretical foundations of these sampling schemes have been provided. Among others, Prioritized Experience Replay appears as a hyperparameter sensitive heuristic, even though it can provide good performance. In this work, we cast the replay buffer sampling problem as an importance sampling one for estimating the gradient. This allows deriving the theoretically optimal sampling distribution, yielding the best theoretical convergence speed.…
Experience replay is crucial for off-policy reinforcement learning (RL) methods. By remembering and reusing the experiences from past different policies, experience replay significantly improves the training efficiency and stability of RL algorithms. Many decision-making problems in practice naturally involve multiple agents and require multi-agent reinforcement learning (MARL) under centralized training decentralized execution paradigm. Nevertheless, existing MARL algorithms often adopt standard experience replay where the transitions are uniformly sampled regardless of their importance. Find…
Freshness-Aware Prioritized Experience Replay for LLM/VLM Reinforcement Learning
King Abdullah University of Science and Technology（卡布尔大学科学与技术大学）；Chinese Academy of Sciences, Institute of Automation（中国科学院自动化研究所）；AI Centre, Department of Computer Science, University College London（伦敦大学学院人工智能中心）；Zhongguancun Institute of Artificial Intelligence（中关村人工智能研究所）
Recently experience replay is widely used in various deep reinforcement learning (RL) algorithms, in this paper we rethink the utility of experience replay. It introduces a new hyper-parameter, the memory buffer size, which needs carefully tuning. However unfortunately the importance of this new hyper-parameter has been underestimated in the community for a long time. In this paper we did a systematic empirical study of experience replay under various function representations. We showcase that a large replay buffer can significantly hurt the performance. Moreover, we propose a simple O(1) meth…
An important component of many Deep Reinforcement Learning algorithms is the Experience Replay which serves as a storage mechanism or memory of made experiences. These experiences are used for training and help the agent to stably find the perfect trajectory through the problem space. The classic Experience Replay however makes only use of the experiences it actually made, but the stored samples bear great potential in form of knowledge about the problem that can be extracted. We present an algorithm that creates synthetic experiences in a nondeterministic discrete environment to assist the le…
MILA, Université de Montréal（MILA，蒙特利尔大学）；Google Brain（谷歌大脑）；DeepMind
Experience replay is central to off-policy algorithms in deep reinforcement learning (RL), but there remain significant gaps in our understanding. We therefore present a systematic and extensive analysis of experience replay in Q-learning methods, focusing on two fundamental properties: the replay capacity and the ratio of learning updates to experience collected (replay ratio). Our additive and ablative studies upend conventional wisdom around experience replay -- greater capacity is found to substantially increase the performance of certain algorithms, while leaving others unaffected. Counte…
Experience replay (ER) is a fundamental component of off-policy deep reinforcement learning (RL). ER recalls experiences from past iterations to compute gradient estimates for the current policy, increasing data-efficiency. However, the accuracy of such updates may deteriorate when the policy diverges from past behaviors and can undermine the performance of ER. Many algorithms mitigate this issue by tuning hyper-parameters to slow down policy changes. An alternative is to actively enforce the similarity between policy and the experiences in the replay memory. We introduce Remember and Forget E…
Experience replay \citep{lin1993reinforcement, mnih2015human} is a widely used technique to achieve efficient use of data and improved performance in RL algorithms. In experience replay, past transitions are stored in a memory buffer and re-used during learning. Various suggestions for sampling schemes from the replay buffer have been suggested in previous works, attempting to optimally choose those experiences which will most contribute to the convergence to an optimal policy. Here, we give some conditions on the replay sampling scheme that will ensure convergence, focusing on the well-known …
Prioritized Experience Replay (PER) is a deep reinforcement learning technique in which agents learn from transitions sampled with non-uniform probability proportionate to their temporal-difference error. We show that any loss function evaluated with non-uniformly sampled data can be transformed into another uniformly sampled loss function with the same expected gradient. Surprisingly, we find in some environments PER can be replaced entirely by this new loss function without impact to empirical performance. Furthermore, this relationship suggests a new branch of improvements to PER by correct…
Hindsight Experience Replay (HER) is a multi-goal reinforcement learning algorithm for sparse reward functions. The algorithm treats every failure as a success for an alternative (virtual) goal that has been achieved in the episode. Virtual goals are randomly selected, irrespective of which are most instructive for the agent. In this paper, we present two improvements over the existing HER algorithm. First, we prioritize virtual goals from which the agent will learn more valuable information. We call this property the instructiveness of the virtual goal and define it by a heuristic measure, wh…
University of Toronto（多伦多大学）；Vector Institute（向量研究所）；Stanford University（斯坦福大学）；Nvidia（英伟达）
The use of past experiences to accelerate temporal difference (TD) learning of value functions, or experience replay, is a key component in deep reinforcement learning. Prioritization or reweighting of important experiences has shown to improve performance of TD learning algorithms.In this work, we propose to reweight experiences based on their likelihood under the stationary distribution of the current policy. Using the corresponding reweighted TD objective, we implicitly encourage small approximation errors on the value function over frequently encountered states. We use a likelihood-free de…
A commonly used heuristic in RL is experience replay (e.g.~\citet{lin1993reinforcement, mnih2015human}), in which a learner stores and re-uses past trajectories as if they were sampled online. In this work, we initiate a rigorous study of this heuristic in the setting of tabular Q-learning. We provide a convergence rate guarantee, and discuss how it compares to the convergence of Q-learning depending on important parameters such as the frequency and number of replay iterations. We also provide theoretical evidence showing when we might expect this heuristic to strictly improve performance, by …
University of Augsburg（奥格斯堡大学）；BSH Home Appliances（博西家用电器公司）
XCS constitutes the most deeply investigated classifier system today. It bears strong potentials and comes with inherent capabilities for mastering a variety of different learning tasks. Besides outstanding successes in various classification and regression tasks, XCS also proved very effective in certain multi-step environments from the domain of reinforcement learning. Especially in the latter domain, recent advances have been mainly driven by algorithms which model their policies based on deep neural networks -- among which the Deep-Q-Network (DQN) is a prominent representative. Experience …
In this paper, we consider cooperative multi-agent reinforcement learning (MARL) with sparse reward. To tackle this problem, we propose a novel method named MASER: MARL with subgoals generated from experience replay buffer. Under the widely-used assumption of centralized training with decentralized execution and consistent Q-value decomposition for MARL, MASER automatically generates proper subgoals for multiple agents from the experience replay buffer by considering both individual Q-value and total Q-value. Then, MASER designs individual intrinsic reward for each agent based on actionable re…
Deep Reinforcement Learning (RL) methods rely on experience replay to approximate the minibatched supervised learning setting; however, unlike supervised learning where access to lots of training data is crucial to generalization, replay-based deep RL appears to struggle in the presence of extraneous data. Recent works have shown that the performance of Deep Q-Network (DQN) degrades when its replay memory becomes too large. This suggests that outdated experiences somehow impact the performance of deep RL, which should not be the case for off-policy methods like DQN. Consequently, we re-examine…
Experience replay allows a reinforcement learning agent to train on samples from a large amount of the most recent experiences. A simple in-RAM experience replay stores these most recent experiences in a list in RAM, and then copies sampled batches to the GPU for training. I moved this list to the GPU, thus creating an in-GPU experience replay, and a training step that no longer has inputs copied from the CPU. I trained an agent to play Super Smash Bros. Melee, using internal game memory values as inputs and outputting controller button presses. A single state in Melee contains 27 floats, so t…
We present a novel technique called Dynamic Experience Replay (DER) that allows Reinforcement Learning (RL) algorithms to use experience replay samples not only from human demonstrations but also successful transitions generated by RL agents during training and therefore improve training efficiency. It can be combined with an arbitrary off-policy RL algorithm, such as DDPG or DQN, and their distributed versions. We build upon Ape-X DDPG and demonstrate our approach on robotic tight-fitting joint assembly tasks, based on force/torque and Cartesian pose observations. In particular, we run experi…
Nanjing University（南京大学）；University of New South Wales（新南威尔士大学）
In this paper, a novel training paradigm inspired by quantum computation is proposed for deep reinforcement learning (DRL) with experience replay. In contrast to traditional experience replay mechanism in DRL, the proposed deep reinforcement learning with quantum-inspired experience replay (DRL-QER) adaptively chooses experiences from the replay buffer according to the complexity and the replayed times of each experience (also called transition), to achieve a balance between exploration and exploitation. In DRL-QER, transitions are first formulated in quantum representations, and then the prep…
Efficient Diversity-based Experience Replay for Deep Reinforcement Learning
School of Computer Science, Wuhan University（武汉大学计算机学院）；State Key Laboratory of Internet of Things for Smart City, University of Macau（澳门大学智能城市物联网国家重点实验室）；School of Artificial Intelligence Shenzhen Polytechnic University（深圳职业技术学院人工智能学院）
University of Pennsylvania（宾夕法尼亚大学）；DeepMind（深度思维）
Continual learning is the problem of learning new tasks or knowledge while protecting old knowledge and ideally generalizing from old experience to learn new tasks faster. Neural networks trained by stochastic gradient descent often degrade on old tasks when trained successively on new tasks with different data distributions. This phenomenon, referred to as catastrophic forgetting, is considered a major hurdle to learning with non-stationary data or sequences of new tasks, and prevents networks from continually accumulating knowledge and skills. We examine this issue in the context of reinforc…
