PyTorchKR Papers

최신 논문

총 1,872편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2203.07171공식 코드 · Papers with Code

Orchestrated Value Mapping for Reinforcement Learning

We present a general convergent class of reinforcement learning algorithms that is founded on two distinct principles: (1) mapping value estimates to a different space using arbitrary functions from a broad class, and (2) linearly decomposing the reward signal into multiple channels. The first principle enables incorporating specific properties into the value estimator that can enhance learning. The second principle, on the other hand, allows for the value function to be represented as a composition of multiple utility functions. This can be leveraged for various purposes, e.g. dealing with highly varying reward scales, incorporating a priori knowledge about the sources of reward, and ensemble learning. Combining the two principles yields a general blueprint for instantiating convergent algorithms by orchestrating diverse mapping functions over multiple reward channels. This blueprint generalizes and subsumes algorithms such as Q-Learning, Log Q-Learning, and Q-Decomposition. In addition, our convergence proof for this general class relaxes certain required assumptions in some of these algorithms. Based on our theory, we discuss several interesting configurations as special cases. Finally, to illustrate the potential of the design space that our theory opens up, we instantiate a particular algorithm and evaluate its performance on the Atari suite.

arXiv:2203.06053

A Machine Learning Approach for Prosumer Management in Intraday Electricity Markets

Prosumer operators are dealing with extensive challenges to participate in short-term electricity markets while taking uncertainties into account. Challenges such as variation in demand, solar energy, wind power, and electricity prices as well as faster response time in intraday electricity markets. Machine learning approaches could resolve these challenges due to their ability to continuous learning of complex relations and providing a real-time response. Such approaches are applicable with presence of the high performance computing and big data. To tackle these challenges, a Markov decision process is proposed and solved with a reinforcement learning algorithm with proper observations and actions employing tabular Q-learning. Trained agent converges to a policy which is similar to the global optimal solution. It increases the prosumer's profit by 13.39% compared to the well-known stochastic optimization approach.

arXiv:2203.04120

Graph-based Reinforcement Learning meets Mixed Integer Programs: An application to 3D robot assembly discovery

Robot assembly discovery is a challenging problem that lives at the intersection of resource allocation and motion planning. The goal is to combine a predefined set of objects to form something new while considering task execution with the robot-in-the-loop. In this work, we tackle the problem of building arbitrary, predefined target structures entirely from scratch using a set of Tetris-like building blocks and a robotic manipulator. Our novel hierarchical approach aims at efficiently decomposing the overall task into three feasible levels that benefit mutually from each other. On the high level, we run a classical mixed-integer program for global optimization of block-type selection and the blocks' final poses to recreate the desired shape. Its output is then exploited to efficiently guide the exploration of an underlying reinforcement learning (RL) policy. This RL policy draws its generalization properties from a flexible graph-based representation that is learned through Q-learning and can be refined with search. Moreover, it accounts for the necessary conditions of structural stability and robotic feasibility that cannot be effectively reflected in the previous layer. Lastly, a grasp and motion planner transforms the desired assembly commands into robot joint movements. We demonstrate our proposed method's performance on a set of competitive simulated RAD environments, showcase real-world transfer, and report performance and robustness gains compared to an unstructured end-to-end approach. Videos are available at https://sites.google.com/view/rl-meets-milp .

arXiv:2203.03417

Scalable multi-agent reinforcement learning for distributed control of residential energy flexibility

This paper proposes a novel scalable type of multi-agent reinforcement learning-based coordination for distributed residential energy. Cooperating agents learn to control the flexibility offered by electric vehicles, space heating and flexible loads in a partially observable stochastic environment. In the standard independent Q-learning approach, the coordination performance of agents under partial observability drops at scale in stochastic environments. Here, the novel combination of learning from off-line convex optimisations on historical data and isolating marginal contributions to total rewards in reward signals increases stability and performance at scale. Using fixed-size Q-tables, prosumers are able to assess their marginal impact on total system objectives without sharing personal data either with each other or with a central coordinator. Case studies are used to assess the fitness of different combinations of exploration sources, reward definitions, and multi-agent learning frameworks. It is demonstrated that the proposed strategies create value at individual and system levels thanks to reductions in the costs of energy imports, losses, distribution network congestion, battery depreciation and greenhouse gas emissions.

arXiv:2203.03003

Offline Deep Reinforcement Learning for Dynamic Pricing of Consumer Credit

We introduce a method for pricing consumer credit using recent advances in offline deep reinforcement learning. This approach relies on a static dataset and requires no assumptions on the functional form of demand. Using both real and synthetic data on consumer credit applications, we demonstrate that our approach using the conservative Q-Learning algorithm is capable of learning an effective personalized pricing policy without any online interaction or price experimentation.

arXiv:2203.02628

Target Network and Truncation Overcome The Deadly Triad in $Q$-Learning

$Q$-learning with function approximation is one of the most empirically successful while theoretically mysterious reinforcement learning (RL) algorithms, and was identified in Sutton (1999) as one of the most important theoretical open problems in the RL community. Even in the basic linear function approximation setting, there are well-known divergent examples. In this work, we show that \textit{target network} and \textit{truncation} together are enough to provably stabilize $Q$-learning with linear function approximation, and we establish the finite-sample guarantees. The result implies an $O(\epsilon^{-2})$ sample complexity up to a function approximation error. Moreover, our results do not require strong assumptions or modifying the problem parameters as in existing literature.

arXiv:2203.01004

Improving the Diversity of Bootstrapped DQN by Replacing Priors With Noise

Q-learning is one of the most well-known Reinforcement Learning algorithms. There have been tremendous efforts to develop this algorithm using neural networks. Bootstrapped Deep Q-Learning Network is amongst them. It utilizes multiple neural network heads to introduce diversity into Q-learning. Diversity can sometimes be viewed as the amount of reasonable moves an agent can take at a given state, analogous to the definition of the exploration ratio in RL. Thus, the performance of Bootstrapped Deep Q-Learning Network is deeply connected with the level of diversity within the algorithm. In the original research, it was pointed out that a random prior could improve the performance of the model. In this article, we further explore the possibility of replacing priors with noise and sample the noise from a Gaussian distribution to introduce more diversity into this algorithm. We conduct our experiment on the Atari benchmark and compare our algorithm to both the original and other related algorithms. The results show that our modification of the Bootstrapped Deep Q-Learning algorithm achieves significantly higher evaluation scores across different types of Atari games. Thus, we conclude that replacing priors with noise can improve Bootstrapped Deep Q-Learning's performance by ensuring the integrity of diversities.

arXiv:2203.00885

A Learning Based Framework for Handling Uncertain Lead Times in Multi-Product Inventory Management

Most existing literature on supply chain and inventory management consider stochastic demand processes with zero or constant lead times. While it is true that in certain niche scenarios, uncertainty in lead times can be ignored, most real-world scenarios exhibit stochasticity in lead times. These random fluctuations can be caused due to uncertainty in arrival of raw materials at the manufacturer's end, delay in transportation, an unforeseen surge in demands, and switching to a different vendor, to name a few. Stochasticity in lead times is known to severely degrade the performance in an inventory management system, and it is only fair to abridge this gap in supply chain system through a principled approach. Motivated by the recently introduced delay-resolved deep Q-learning (DRDQN) algorithm, this paper develops a reinforcement learning based paradigm for handling uncertainty in lead times (\emph{action delay}). Through empirical evaluations, it is further shown that the inventory management with uncertain lead times is not only equivalent to that of delay in information sharing across multiple echelons (\emph{observation delay}), a model trained to handle one kind of delay is capable to handle delays of another kind without requiring to be retrained. Finally, we apply the delay-resolved framework to scenarios comprising of multiple products subjected to stochasticity in lead times, and elucidate how the delay-resolved framework negates the effect of any delay to achieve near-optimal performance.

arXiv:2202.13187

Whittle Index based Q-Learning for Wireless Edge Caching with Linear Function Approximation

We consider the problem of content caching at the wireless edge to serve a set of end users via unreliable wireless channels so as to minimize the average latency experienced by end users due to the constrained wireless edge cache capacity. We formulate this problem as a Markov decision process, or more specifically a restless multi-armed bandit problem, which is provably hard to solve. We begin by investigating a discounted counterpart, and prove that it admits an optimal policy of the threshold-type. We then show that this result also holds for average latency problem. Using this structural result, we establish the indexability of our problem, and employ the Whittle index policy to minimize average latency. Since system parameters such as content request rates and wireless channel conditions are often unknown and time-varying, we further develop a model-free reinforcement learning algorithm dubbed as Q^{+}-Whittle that relies on Whittle index policy. However, Q^{+}-Whittle requires to store the Q-function values for all state-action pairs, the number of which can be extremely large for wireless edge caching. To this end, we approximate the Q-function by a parameterized function class with a much smaller dimension, and further design a Q^{+}-Whittle algorithm with linear function approximation, which is called Q^{+}-Whittle-LFA. We provide a finite-time bound on the mean-square error of Q^{+}-Whittle-LFA. Simulation results using real traces demonstrate that Q^{+}-Whittle-LFA yields excellent empirical performance.

arXiv:2202.13890

Pessimistic Q-Learning for Offline Reinforcement Learning: Towards Optimal Sample Complexity

Offline or batch reinforcement learning seeks to learn a near-optimal policy using history data without active exploration of the environment. To counter the insufficient coverage and sample scarcity of many offline datasets, the principle of pessimism has been recently introduced to mitigate high bias of the estimated values. While pessimistic variants of model-based algorithms (e.g., value iteration with lower confidence bounds) have been theoretically investigated, their model-free counterparts -- which do not require explicit model estimation -- have not been adequately studied, especially in terms of sample efficiency. To address this inadequacy, we study a pessimistic variant of Q-learning in the context of finite-horizon Markov decision processes, and characterize its sample complexity under the single-policy concentrability assumption which does not require the full coverage of the state-action space. In addition, a variance-reduced pessimistic Q-learning algorithm is proposed to achieve near-optimal sample complexity. Altogether, this work highlights the efficiency of model-free algorithms in offline RL when used in conjunction with pessimism and variance reduction.

arXiv:2202.10338

UAV Base Station Trajectory Optimization Based on Reinforcement Learning in Post-disaster Search and Rescue Operations

Because of disaster, terrestrial base stations (TBS) would be partly crashed. Some user equipments (UE) would be unserved. Deploying unmanned aerial vehicles (UAV) as aerial base stations is a method to cover UEs quickly. But existing methods solely refer to the coverage of UAVs. In those scenarios, they focus on the deployment of UAVs in the post-disaster area where all TBSs do not work any longer. There is limited research about the combination of available TBSs and UAVs. We propose the method to deploy UAVs cooperating with available TBSs as aerial base stations. And improve the coverage by reinforcement learning. Besides, in the experiments, we cluster UEs with balanced iterative reducing and clustering using hierarchies (BIRCH) at first. Finally, achieve base stations' better coverage to UEs through Q-learning.

arXiv:2202.09722

PooL: Pheromone-inspired Communication Framework forLarge Scale Multi-Agent Reinforcement Learning

Being difficult to scale poses great problems in multi-agent coordination. Multi-agent Reinforcement Learning (MARL) algorithms applied in small-scale multi-agent systems are hard to extend to large-scale ones because the latter is far more dynamic and the number of interactions increases exponentially with the growing number of agents. Some swarm intelligence algorithms simulate the release and utilization mechanism of pheromones to control large-scale agent coordination. Inspired by such algorithms, \textbf{PooL}, an \textbf{p}her\textbf{o}m\textbf{o}ne-based indirect communication framework applied to large scale multi-agent reinforcement \textbf{l}earning is proposed in order to solve the large-scale multi-agent coordination problem. Pheromones released by agents of PooL are defined as outputs of most reinforcement learning algorithms, which reflect agents' views of the current environment. The pheromone update mechanism can efficiently organize the information of all agents and simplify the complex interactions among agents into low-dimensional representations. Pheromones perceived by agents can be regarded as a summary of the views of nearby agents which can better reflect the real situation of the environment. Q-Learning is taken as our base model to implement PooL and PooL is evaluated in various large-scale cooperative environments. Experiments show agents can capture effective information through PooL and achieve higher rewards than other state-of-arts methods with lower communication costs.

arXiv:2202.10019

Autonomous Warehouse Robot using Deep Q-Learning

In warehouses, specialized agents need to navigate, avoid obstacles and maximize the use of space in the warehouse environment. Due to the unpredictability of these environments, reinforcement learning approaches can be applied to complete these tasks. In this paper, we propose using Deep Reinforcement Learning (DRL) to address the robot navigation and obstacle avoidance problem and traditional Q-learning with minor variations to maximize the use of space for product placement. We first investigate the problem for the single robot case. Next, based on the single robot model, we extend our system to the multi-robot case. We use a strategic variation of Q-tables to perform multi-agent Q-learning. We successfully test the performance of our model in a 2D simulation environment for both the single and multi-robot cases.

논문 ID: publication/3491416

Variation-resistant Q-learning: Controlling and Utilizing Estimation Bias in Reinforcement Learning for Better Performance

Q-learning is a reinforcement learning algorithm that has overestimation bias, because it learns the optimal action values by using a target that maximizes over uncertain action-value estimates. Although the overestima-tion bias of Q-learning is generally considered harmful, a recent study suggests that it could be either harmful or helpful depending on the reinforcement learning problem. In this paper, we propose a new Q-learning variant, called Variation-resistant Q-learning, to control and utilize estimation bias for better performance. Firstly, we present the tabular version of the algorithm and mathematically prove its convergence. Secondly, we combine the algorithm with function approximation. Finally, we present empirical results from three different experiments, in which we compared the performance of Variation-resistant Q-learning, Q-learning, and Double Q-learning. The empirical results show that Variation-resistant Q-learning can control and utilize estimation bias for better performance in the experimental tasks.

arXiv:2202.05947

Artificial Intelligence and Auction Design

Motivated by online advertising auctions, we study auction design in repeated auctions played by simple Artificial Intelligence algorithms (Q-learning). We find that first-price auctions with no additional feedback lead to tacit-collusive outcomes (bids lower than values), while second-price auctions do not. We show that the difference is driven by the incentive in first-price auctions to outbid opponents by just one bid increment. This facilitates re-coordination on low bids after a phase of experimentation. We also show that providing information about lowest bid to win, as introduced by Google at the time of switch to first-price auctions, increases competitiveness of auctions.

arXiv:2202.06356공식 코드 · Papers with Code

Goal Recognition as Reinforcement Learning

Most approaches for goal recognition rely on specifications of the possible dynamics of the actor in the environment when pursuing a goal. These specifications suffer from two key issues. First, encoding these dynamics requires careful design by a domain expert, which is often not robust to noise at recognition time. Second, existing approaches often need costly real-time computations to reason about the likelihood of each potential goal. In this paper, we develop a framework that combines model-free reinforcement learning and goal recognition to alleviate the need for careful, manual domain design, and the need for costly online executions. This framework consists of two main stages: Offline learning of policies or utility functions for each potential goal, and online inference. We provide a first instance of this framework using tabular Q-learning for the learning stage, as well as three measures that can be used to perform the inference stage. The resulting instantiation achieves state-of-the-art performance against goal recognizers on standard evaluation domains and superior performance in noisy environments.

arXiv:2202.05404

Regularized Q-learning

Q-learning is widely used algorithm in reinforcement learning community. Under the lookup table setting, its convergence is well established. However, its behavior is known to be unstable with the linear function approximation case. This paper develops a new Q-learning algorithm that converges when linear function approximation is used. We prove that simply adding an appropriate regularization term ensures convergence of the algorithm. We prove its stability using a recent analysis tool based on switching system models. Moreover, we experimentally show that it converges in environments where Q-learning with linear function approximation has known to diverge. We also provide an error bound on the solution where the algorithm converges.

arXiv:2202.04709

Transferred Q-learning

We consider $Q$-learning with knowledge transfer, using samples from a target reinforcement learning (RL) task as well as source samples from different but related RL tasks. We propose transfer learning algorithms for both batch and online $Q$-learning with offline source studies. The proposed transferred $Q$-learning algorithm contains a novel re-targeting step that enables vertical information-cascading along multiple steps in an RL task, besides the usual horizontal information-gathering as transfer learning (TL) for supervised learning. We establish the first theoretical justifications of TL in RL tasks by showing a faster rate of convergence of the $Q$ function estimation in the offline RL transfer, and a lower regret bound in the offline-to-online RL transfer under certain similarity assumptions. Empirical evidences from both synthetic and real datasets are presented to back up the proposed algorithm and our theoretical results.

arXiv:2202.04224

Intelligent Autonomous Intersection Management

Connected Autonomous Vehicles will make autonomous intersection management a reality replacing traditional traffic signal control. Autonomous intersection management requires time and speed adjustment of vehicles arriving at an intersection for collision-free passing through the intersection. Due to its computational complexity, this problem has been studied only when vehicle arrival times towards the vicinity of the intersection are known beforehand, which limits the applicability of these solutions for real-time deployment. To solve the real-time autonomous traffic intersection management problem, we propose a reinforcement learning (RL) based multiagent architecture and a novel RL algorithm coined multi-discount Q-learning. In multi-discount Q-learning, we introduce a simple yet effective way to solve a Markov Decision Process by preserving both short-term and long-term goals, which is crucial for collision-free speed control. Our empirical results show that our RL-based multiagent solution can achieve near-optimal performance efficiently when minimizing the travel time through an intersection.

arXiv:2202.03600

Multiple Correlated Jammers Nullification using LSTM-based Deep Dueling Neural Network

Suppressing the deliberate interference for wireless networks is critical to guarantee a reliable communication link. However, nullifying the jamming signals can be problematic when the correlations between transmitted jamming signals are deliberately varied over time. Specifically, recent studies reveal that by deliberately varying the correlations among jamming signals, attackers can effectively vary the jamming channels and thus their nullspace, even when the physical channels remain unchanged. That makes the beam-forming matrix derived from the nullspace of the jamming channels unable to suppress the jamming signals. Most existing solutions only consider unchanged correlations or heuristically adapt to the time-varying correlation problem by continuously monitoring the residual jamming signals before updating the beam-forming matrix. In this paper, we systematically formulate the optimization problem of the nullspace estimation and data transmission phases. Even ignoring the unknown strategy of the jammers and the challenging nullspace estimation process, the resulting problem is an integer programming problem, hence intractable to obtain its optimal solution. To tackle it and address the unknown strategy of the jammer, we reformulate the problem using a partially observable semi-Markov decision process (POSMDP) and then design a deep dueling Q-learning based framework to tune the duration of the nullspace estimation and data transmission phases. Extensive simulations demonstrate that the proposed techniques effectively deal with jamming signals whose correlations vary over time, and the range of correlations is unknown. Especially, our techniques do not require continuous monitoring of the residual jamming signals (after the nullification process) before updating the beam-forming matrix. As such, the system is more spectral-efficient and has a lower outage probability.