PyTorchKR Papers

최신 논문

총 1,872편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2306.14489

Decentralized Multi-Robot Formation Control Using Reinforcement Learning

This paper presents a decentralized leader-follower multi-robot formation control based on a reinforcement learning (RL) algorithm applied to a swarm of small educational Sphero robots. Since the basic Q-learning method is known to require large memory resources for Q-tables, this work implements the Double Deep Q-Network (DDQN) algorithm, which has achieved excellent results in many robotic problems. To enhance the system behavior, we trained two different DDQN models, one for reaching the formation and the other for maintaining it. The models use a discrete set of robot motions (actions) to adapt the continuous nonlinear system to the discrete nature of RL. The presented approach has been tested in simulation and real experiments which show that the multi-robot system can achieve and maintain a stable formation without the need for complex mathematical models and nonlinear control laws.

arXiv:2306.11918

Adaptive Ensemble Q-learning: Minimizing Estimation Bias via Error Feedback

The ensemble method is a promising way to mitigate the overestimation issue in Q-learning, where multiple function approximators are used to estimate the action values. It is known that the estimation bias hinges heavily on the ensemble size (i.e., the number of Q-function approximators used in the target), and that determining the `right' ensemble size is highly nontrivial, because of the time-varying nature of the function approximation errors during the learning process. To tackle this challenge, we first derive an upper bound and a lower bound on the estimation bias, based on which the ensemble size is adapted to drive the bias to be nearly zero, thereby coping with the impact of the time-varying approximation errors accordingly. Motivated by the theoretic findings, we advocate that the ensemble method can be combined with Model Identification Adaptive Control (MIAC) for effective ensemble size adaptation. Specifically, we devise Adaptive Ensemble Q-learning (AdaEQ), a generalized ensemble method with two key steps: (a) approximation error characterization which serves as the feedback for flexibly controlling the ensemble size, and (b) ensemble size adaptation tailored towards minimizing the estimation bias. Extensive experiments are carried out to show that AdaEQ can improve the learning performance than the existing methods for the MuJoCo benchmark.

arXiv:2306.10216

Vanishing Bias Heuristic-guided Reinforcement Learning Algorithm

Reinforcement Learning has achieved tremendous success in the many Atari games. In this paper we explored with the lunar lander environment and implemented classical methods including Q-Learning, SARSA, MC as well as tiling coding. We also implemented Neural Network based methods including DQN, Double DQN, Clipped DQN. On top of these, we proposed a new algorithm called Heuristic RL which utilizes heuristic to guide the early stage training while alleviating the introduced human bias. Our experiments showed promising results for our proposed methods in the lunar lander environment.

arXiv:2306.10071공식 코드 · Papers with Code

Joint Path planning and Power Allocation of a Cellular-Connected UAV using Apprenticeship Learning via Deep Inverse Reinforcement Learning

This paper investigates an interference-aware joint path planning and power allocation mechanism for a cellular-connected unmanned aerial vehicle (UAV) in a sparse suburban environment. The UAV's goal is to fly from an initial point and reach a destination point by moving along the cells to guarantee the required quality of service (QoS). In particular, the UAV aims to maximize its uplink throughput and minimize the level of interference to the ground user equipment (UEs) connected to the neighbor cellular BSs, considering the shortest path and flight resource limitation. Expert knowledge is used to experience the scenario and define the desired behavior for the sake of the agent (i.e., UAV) training. To solve the problem, an apprenticeship learning method is utilized via inverse reinforcement learning (IRL) based on both Q-learning and deep reinforcement learning (DRL). The performance of this method is compared to learning from a demonstration technique called behavioral cloning (BC) using a supervised learning approach. Simulation and numerical results show that the proposed approach can achieve expert-level performance. We also demonstrate that, unlike the BC technique, the performance of our proposed approach does not degrade in unseen situations.

arXiv:2306.11217

Autonomous Driving with Deep Reinforcement Learning in CARLA Simulation

Nowadays, autonomous vehicles are gaining traction due to their numerous potential applications in resolving a variety of other real-world challenges. However, developing autonomous vehicles need huge amount of training and testing before deploying it to real world. While the field of reinforcement learning (RL) has evolved into a powerful learning framework to the development of deep representation learning, and it is now capable of learning complicated policies in high-dimensional environments like in autonomous vehicles. In this regard, we make an effort, using Deep Q-Learning, to discover a method by which an autonomous car may maintain its lane at top speed while avoiding other vehicles. After that, we used CARLA simulation environment to test and verify our newly acquired policy based on the problem formulation.

arXiv:2306.09915

Model-based versus model-free feeding control and water quality monitoring for fish growth tracking in aquaculture systems

The high concentration level of the environmental factors, such as a high ammonia concentration and pH level, affect the water quality, affecting fish's survival and mass death. Therefore, there is a critical need to develop control strategies to determine optimal, efficient, and reliable feeding and water quality monitoring processes. In this paper, we revisit the representative fish growth model describing the total biomass change by incorporating the fish population density and mortality. Since the measurement data of the total biomass and population from the aquaculture systems are limited and difficult to obtain, we validate the new dynamic population model with the individual fish growth data for tracking control purposes. We specifically focus on relative feeding as a manipulated variable to design traditional and optimal control to track the desired weight reference within the sub-optimal temperature and dissolved oxygen profiles under different levels of unionized ammonia exposure. Then, we propose a Q-learning approach that learns an optimal feeding control policy from the simulated data of the fish growth weight trajectories while managing the ammonia effects. The proposed Q-learning feeding control prevents fish mortality and achieves good tracking errors of the fish weight under the different levels of unionized ammonia. However, it maintains a relative food consumption that potentially underfeeds the fish. Finally, we propose an optimal algorithm that optimizes the feeding and water quality of the dynamic fish population growth process. We also show that the model predictive control decreases fish mortality and reduces food consumption in all different cases of unionized ammonia exposure.

arXiv:2306.09437

Algorithmic Collusion in Auctions: Evidence from Controlled Laboratory Experiments

Algorithms are increasingly being used to automate participation in online markets. Banchio and Skrzypacz (2022) demonstrate how exploration under identical valuation in first-price auctions may lead to spontaneous coupling into sub-competitive bidding. However, it is an open question if these findings extend to affiliated values, optimal exploration, and specifically which algorithmic details play a role in facilitating algorithmic collusion. This paper contributes to the literature by generating robust stylized facts to cover these gaps. I conduct a set of fully randomized experiments in a controlled laboratory setup and apply double machine learning to estimate granular conditional treatment effects of auction design on seller revenues. I find that first-price auctions lead to lower seller revenues and higher seller regret under identical values, affiliated values, and under both Q-learning and Bandits. There is more possibility of such tacit collusion under fewer bidders, Boltzmann exploration, asynchronous updating, and longer episodes; while high reserve prices can offset this. This evidence suggests that programmatic auctions, e.g. the Google Ad Exchange, which depend on first-price auctions, might be susceptible to coordinated bid suppression and significant revenue losses.

arXiv:2306.09526

Residual Q-Learning: Offline and Online Policy Customization without Value

Imitation Learning (IL) is a widely used framework for learning imitative behavior from demonstrations. It is especially appealing for solving complex real-world tasks where handcrafting reward function is difficult, or when the goal is to mimic human expert behavior. However, the learned imitative policy can only follow the behavior in the demonstration. When applying the imitative policy, we may need to customize the policy behavior to meet different requirements coming from diverse downstream tasks. Meanwhile, we still want the customized policy to maintain its imitative nature. To this end, we formulate a new problem setting called policy customization. It defines the learning task as training a policy that inherits the characteristics of the prior policy while satisfying some additional requirements imposed by a target downstream task. We propose a novel and principled approach to interpret and determine the trade-off between the two task objectives. Specifically, we formulate the customization problem as a Markov Decision Process (MDP) with a reward function that combines 1) the inherent reward of the demonstration; and 2) the add-on reward specified by the downstream task. We propose a novel framework, Residual Q-learning, which can solve the formulated MDP by leveraging the prior policy without knowing the inherent reward or value function of the prior policy. We derive a family of residual Q-learning algorithms that can realize offline and online policy customization, and show that the proposed algorithms can effectively accomplish policy customization tasks in various environments. Demo videos and code are available on our website: https://sites.google.com/view/residualq-learning.

arXiv:2306.09273

Privacy Risks in Reinforcement Learning for Household Robots

The prominence of embodied Artificial Intelligence (AI), which empowers robots to navigate, perceive, and engage within virtual environments, has attracted significant attention, owing to the remarkable advances in computer vision and large language models. Privacy emerges as a pivotal concern within the realm of embodied AI, as the robot accesses substantial personal information. However, the issue of privacy leakage in embodied AI tasks, particularly concerning reinforcement learning algorithms, has not received adequate consideration in research. This paper aims to address this gap by proposing an attack on the training process of the value-based algorithm and the gradient-based algorithm, utilizing gradient inversion to reconstruct states, actions, and supervisory signals. The choice of using gradients for the attack is motivated by the fact that commonly employed federated learning techniques solely utilize gradients computed based on private user data to optimize models, without storing or transmitting the data to public servers. Nevertheless, these gradients contain sufficient information to potentially expose private data. To validate our approach, we conducted experiments on the AI2THOR simulator and evaluated our algorithm on active perception, a prevalent task in embodied AI. The experimental results demonstrate the effectiveness of our method in successfully reconstructing all information from the data in 120 room layouts. Check our website for videos.

arXiv:2306.08044

Pruning the Way to Reliable Policies: A Multi-Objective Deep Q-Learning Approach to Critical Care

Medical treatments often involve a sequence of decisions, each informed by previous outcomes. This process closely aligns with reinforcement learning (RL), a framework for optimizing sequential decisions to maximize cumulative rewards under unknown dynamics. While RL shows promise for creating data-driven treatment plans, its application in medical contexts is challenging due to the frequent need to use sparse rewards, primarily defined based on mortality outcomes. This sparsity can reduce the stability of offline estimates, posing a significant hurdle in fully utilizing RL for medical decision-making. We introduce a deep Q-learning approach to obtain more reliable critical care policies by integrating relevant but noisy frequently measured biomarker signals into the reward specification without compromising the optimization of the main outcome. Our method prunes the action space based on all available rewards before training a final model on the sparse main reward. This approach minimizes potential distortions of the main objective while extracting valuable information from intermediate signals to guide learning. We evaluate our method in off-policy and offline settings using simulated environments and real health records from intensive care units. Our empirical results demonstrate that our method outperforms common offline RL methods such as conservative Q-learning and batch-constrained deep Q-learning. By disentangling sparse rewards and frequently measured reward proxies through action pruning, our work represents a step towards developing reliable policies that effectively harness the wealth of available information in data-intensive critical care environments.

arXiv:2306.05700

Finite-Time Analysis of Minimax Q-Learning for Two-Player Zero-Sum Markov Games: Switching System Approach

The objective of this paper is to investigate the finite-time analysis of a Q-learning algorithm applied to two-player zero-sum Markov games. Specifically, we establish a finite-time analysis of both the minimax Q-learning algorithm and the corresponding value iteration method. To enhance the analysis of both value iteration and Q-learning, we employ the switching system model of minimax Q-learning and the associated value iteration. This approach provides further insights into minimax Q-learning and facilitates a more straightforward and insightful convergence analysis. We anticipate that the introduction of these additional insights has the potential to uncover novel connections and foster collaboration between concepts in the fields of control theory and reinforcement learning communities.

arXiv:2306.05991

Approximate information state based convergence analysis of recurrent Q-learning

In spite of the large literature on reinforcement learning (RL) algorithms for partially observable Markov decision processes (POMDPs), a complete theoretical understanding is still lacking. In a partially observable setting, the history of data available to the agent increases over time so most practical algorithms either truncate the history to a finite window or compress it using a recurrent neural network leading to an agent state that is non-Markovian. In this paper, it is shown that in spite of the lack of the Markov property, recurrent Q-learning (RQL) converges in the tabular setting. Moreover, it is shown that the quality of the converged limit depends on the quality of the representation which is quantified in terms of what is known as an approximate information state (AIS). Based on this characterization of the approximation error, a variant of RQL with AIS losses is presented. This variant performs better than a strong baseline for RQL that does not use AIS losses. It is demonstrated that there is a strong correlation between the performance of RQL over time and the loss associated with the AIS representation.

arXiv:2306.05053

Active Inference in Hebbian Learning Networks

This work studies how brain-inspired neural ensembles equipped with local Hebbian plasticity can perform active inference (AIF) in order to control dynamical agents. A generative model capturing the environment dynamics is learned by a network composed of two distinct Hebbian ensembles: a posterior network, which infers latent states given the observations, and a state transition network, which predicts the next expected latent state given current state-action pairs. Experimental studies are conducted using the Mountain Car environment from the OpenAI gym suite, to study the effect of the various Hebbian network parameters on the task performance. It is shown that the proposed Hebbian AIF approach outperforms the use of Q-learning, while not requiring any replay buffer, as in typical reinforcement learning systems. These results motivate further investigations of Hebbian learning for the design of AIF networks that can learn environment dynamics without the need for revisiting past buffered experiences.

arXiv:2306.03985공식 코드 · Papers with Code

Agent Performing Autonomous Stock Trading under Good and Bad Situations

Stock trading is one of the popular ways for financial management. However, the market and the environment of economy is unstable and usually not predictable. Furthermore, engaging in stock trading requires time and effort to analyze, create strategies, and make decisions. It would be convenient and effective if an agent could assist or even do the task of analyzing and modeling the past data and then generate a strategy for autonomous trading. Recently, reinforcement learning has been shown to be robust in various tasks that involve achieving a goal with a decision making strategy based on time-series data. In this project, we have developed a pipeline that simulates the stock trading environment and have trained an agent to automate the stock trading process with deep reinforcement learning methods, including deep Q-learning, deep SARSA, and the policy gradient method. We evaluate our platform during relatively good (before 2021) and bad (2021 - 2022) situations. The stocks we've evaluated on including Google, Apple, Tesla, Meta, Microsoft, and IBM. These stocks are among the popular ones, and the changes in trends are representative in terms of having good and bad situations. We showed that before 2021, the three reinforcement methods we have tried always provide promising profit returns with total annual rates around $70\%$ to $90\%$, while maintain a positive profit return after 2021 with total annual rates around 2% to 7%.

arXiv:2306.03951

Reinforcement Learning-Based Control of CrazyFlie 2.X Quadrotor

The objective of the project is to explore synergies between classical control algorithms such as PID and contemporary reinforcement learning algorithms to come up with a pragmatic control mechanism to control the CrazyFlie 2.X quadrotor. The primary objective would be performing PID tuning using reinforcement learning strategies. The secondary objective is to leverage the learnings from the first task to implement control for navigation by integrating with the lighthouse positioning system. Two approaches are considered for navigation, a discrete navigation problem using Deep Q-Learning with finite predefined motion primitives, and deep reinforcement learning for a continuous navigation approach. Simulations for RL training will be performed on gym-pybullet-drones, an open-source gym-based environment for reinforcement learning, and the RL implementations are provided by stable-baselines3

arXiv:2306.01451

Deep Q-Learning versus Proximal Policy Optimization: Performance Comparison in a Material Sorting Task

This paper presents a comparison between two well-known deep Reinforcement Learning (RL) algorithms: Deep Q-Learning (DQN) and Proximal Policy Optimization (PPO) in a simulated production system. We utilize a Petri Net (PN)-based simulation environment, which was previously proposed in related work. The performance of the two algorithms is compared based on several evaluation metrics, including average percentage of correctly assembled and sorted products, average episode length, and percentage of successful episodes. The results show that PPO outperforms DQN in terms of all evaluation metrics. The study highlights the advantages of policy-based algorithms in problems with high-dimensional state and action spaces. The study contributes to the field of deep RL in context of production systems by providing insights into the effectiveness of different algorithms and their suitability for different tasks.

arXiv:2306.00867

IQL-TD-MPC: Implicit Q-Learning for Hierarchical Model Predictive Control

Model-based reinforcement learning (RL) has shown great promise due to its sample efficiency, but still struggles with long-horizon sparse-reward tasks, especially in offline settings where the agent learns from a fixed dataset. We hypothesize that model-based RL agents struggle in these environments due to a lack of long-term planning capabilities, and that planning in a temporally abstract model of the environment can alleviate this issue. In this paper, we make two key contributions: 1) we introduce an offline model-based RL algorithm, IQL-TD-MPC, that extends the state-of-the-art Temporal Difference Learning for Model Predictive Control (TD-MPC) with Implicit Q-Learning (IQL); 2) we propose to use IQL-TD-MPC as a Manager in a hierarchical setting with any off-the-shelf offline RL algorithm as a Worker. More specifically, we pre-train a temporally abstract IQL-TD-MPC Manager to predict "intent embeddings", which roughly correspond to subgoals, via planning. We empirically show that augmenting state representations with intent embeddings generated by an IQL-TD-MPC manager significantly improves off-the-shelf offline RL agents' performance on some of the most challenging D4RL benchmark tasks. For instance, the offline RL algorithms AWAC, TD3-BC, DT, and CQL all get zero or near-zero normalized evaluation scores on the medium and large antmaze tasks, while our modification gives an average score over 40.

arXiv:2305.18443

Off-Policy RL Algorithms Can be Sample-Efficient for Continuous Control via Sample Multiple Reuse

Sample efficiency is one of the most critical issues for online reinforcement learning (RL). Existing methods achieve higher sample efficiency by adopting model-based methods, Q-ensemble, or better exploration mechanisms. We, instead, propose to train an off-policy RL agent via updating on a fixed sampled batch multiple times, thus reusing these samples and better exploiting them within a single optimization loop. We name our method sample multiple reuse (SMR). We theoretically show the properties of Q-learning with SMR, e.g., convergence. Furthermore, we incorporate SMR with off-the-shelf off-policy RL algorithms and conduct experiments on a variety of continuous control benchmarks. Empirical results show that SMR significantly boosts the sample efficiency of the base methods across most of the evaluated tasks without any hyperparameter tuning or additional tricks.

arXiv:2305.18420

Sample Complexity of Variance-reduced Distributionally Robust Q-learning

Dynamic decision-making under distributional shifts is of fundamental interest in theory and applications of reinforcement learning: The distribution of the environment in which the data is collected can differ from that of the environment in which the model is deployed. This paper presents two novel model-free algorithms, namely the distributionally robust Q-learning and its variance-reduced counterpart, that can effectively learn a robust policy despite distributional shifts. These algorithms are designed to efficiently approximate the $q$-function of an infinite-horizon $\gamma$-discounted robust Markov decision process with Kullback-Leibler ambiguity set to an entry-wise $\epsilon$-degree of precision. Further, the variance-reduced distributionally robust Q-learning combines the synchronous Q-learning with variance-reduction techniques to enhance its performance. Consequently, we establish that it attains a minimax sample complexity upper bound of $\tilde O(|\mathbf{S}||\mathbf{A}|(1-\gamma)^{-4}\epsilon^{-2})$, where $\mathbf{S}$ and $\mathbf{A}$ denote the state and action spaces. This is the first complexity result that is independent of the ambiguity size $\delta$, thereby providing new complexity theoretic insights. Additionally, a series of numerical experiments confirm the theoretical findings and the efficiency of the algorithms in handling distributional shifts.

arXiv:2305.18161

VA-learning as a more efficient alternative to Q-learning

In reinforcement learning, the advantage function is critical for policy improvement, but is often extracted from a learned Q-function. A natural question is: Why not learn the advantage function directly? In this work, we introduce VA-learning, which directly learns advantage function and value function using bootstrapping, without explicit reference to Q-functions. VA-learning learns off-policy and enjoys similar theoretical guarantees as Q-learning. Thanks to the direct learning of advantage function and value function, VA-learning improves the sample efficiency over Q-learning both in tabular implementations and deep RL agents on Atari-57 games. We also identify a close connection between VA-learning and the dueling architecture, which partially explains why a simple architectural change to DQN agents tends to improve performance.