PyTorchKR Papers

최신 논문

총 13,741편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2404.18713공식 코드 · Papers with Code

Task and Domain Adaptive Reinforcement Learning for Robot Control

Deep reinforcement learning (DRL) has shown remarkable success in simulation domains, yet its application in designing robot controllers remains limited, due to its single-task orientation and insufficient adaptability to environmental changes. To overcome these limitations, we present a novel adaptive agent that leverages transfer learning techniques to dynamically adapt policy in response to different tasks and environmental conditions. The approach is validated through the blimp control challenge, where multitasking capabilities and environmental adaptability are essential. The agent is trained using a custom, highly parallelized simulator built on IsaacGym. We perform zero-shot transfer to fly the blimp in the real world to solve various tasks. We share our code at https://github.com/robot-perception-group/adaptive_agent.

arXiv:2404.16879

Learning Control Barrier Functions and their application in Reinforcement Learning: A Survey

Reinforcement learning is a powerful technique for developing new robot behaviors. However, typical lack of safety guarantees constitutes a hurdle for its practical application on real robots. To address this issue, safe reinforcement learning aims to incorporate safety considerations, enabling faster transfer to real robots and facilitating lifelong learning. One promising approach within safe reinforcement learning is the use of control barrier functions. These functions provide a framework to ensure that the system remains in a safe state during the learning process. However, synthesizing control barrier functions is not straightforward and often requires ample domain knowledge. This challenge motivates the exploration of data-driven methods for automatically defining control barrier functions, which is highly appealing. We conduct a comprehensive review of the existing literature on safe reinforcement learning using control barrier functions. Additionally, we investigate various techniques for automatically learning the Control Barrier Functions, aiming to enhance the safety and efficacy of Reinforcement Learning in practical robot applications.

arXiv:2404.14713

Enhancing High-Speed Cruising Performance of Autonomous Vehicles through Integrated Deep Reinforcement Learning Framework

High-speed cruising scenarios with mixed traffic greatly challenge the road safety of autonomous vehicles (AVs). Unlike existing works that only look at fundamental modules in isolation, this work enhances AV safety in mixed-traffic high-speed cruising scenarios by proposing an integrated framework that synthesizes three fundamental modules, i.e., behavioral decision-making, path-planning, and motion-control modules. Considering that the integrated framework would increase the system complexity, a bootstrapped deep Q-Network (DQN) is employed to enhance the deep exploration of the reinforcement learning method and achieve adaptive decision making of AVs. Moreover, to make AV behavior understandable by surrounding HDVs to prevent unexpected operations caused by misinterpretations, we derive an inverse reinforcement learning (IRL) approach to learn the reward function of skilled drivers for the path planning of lane-changing maneuvers. Such a design enables AVs to achieve a human-like tradeoff between multi-performance requirements. Simulations demonstrate that the proposed integrated framework can guide AVs to take safe actions while guaranteeing high-speed cruising performance.

arXiv:2404.16767공식 코드 · Papers with Code

REBEL: Reinforcement Learning via Regressing Relative Rewards

While originally developed for continuous control problems, Proximal Policy Optimization (PPO) has emerged as the work-horse of a variety of reinforcement learning (RL) applications, including the fine-tuning of generative models. Unfortunately, PPO requires multiple heuristics to enable stable convergence (e.g. value networks, clipping), and is notorious for its sensitivity to the precise implementation of these components. In response, we take a step back and ask what a minimalist RL algorithm for the era of generative models would look like. We propose REBEL, an algorithm that cleanly reduces the problem of policy optimization to regressing the relative reward between two completions to a prompt in terms of the policy, enabling strikingly lightweight implementation. In theory, we prove that fundamental RL algorithms like Natural Policy Gradient can be seen as variants of REBEL, which allows us to match the strongest known theoretical guarantees in terms of convergence and sample complexity in the RL literature. REBEL can also cleanly incorporate offline data and be extended to handle the intransitive preferences we frequently see in practice. Empirically, we find that REBEL provides a unified approach to language modeling and image generation with stronger or similar performance as PPO and DPO, all while being simpler to implement and more computationally efficient than PPO. When fine-tuning Llama-3-8B-Instruct, REBEL achieves strong performance in AlpacaEval 2.0, MT-Bench, and Open LLM Leaderboard.

arXiv:2404.16300공식 코드 · Papers with Code

Reinforcement Learning with Generative Models for Compact Support Sets

Foundation models contain a wealth of information from their vast number of training samples. However, most prior arts fail to extract this information in a precise and efficient way for small sample sizes. In this work, we propose a framework utilizing reinforcement learning as a control for foundation models, allowing for the granular generation of small, focused synthetic support sets to augment the performance of neural network models on real data classification tasks. We first allow a reinforcement learning agent access to a novel context based dictionary; the agent then uses this dictionary with a novel prompt structure to form and optimize prompts as inputs to generative models, receiving feedback based on a reward function combining the change in validation accuracy and entropy. A support set is formed this way over several exploration steps. Our framework produced excellent results, increasing classification accuracy by significant margins for no additional labelling or data cost.

arXiv:2404.16468공식 코드 · Papers with Code

A Dual Perspective of Reinforcement Learning for Imposing Policy Constraints

Model-free reinforcement learning methods lack an inherent mechanism to impose behavioural constraints on the trained policies. Although certain extensions exist, they remain limited to specific types of constraints, such as value constraints with additional reward signals or visitation density constraints. In this work we unify these existing techniques and bridge the gap with classical optimization and control theory, using a generic primal-dual framework for value-based and actor-critic reinforcement learning methods. The obtained dual formulations turn out to be especially useful for imposing additional constraints on the learned policy, as an intrinsic relationship between such dual constraints (or regularization terms) and reward modifications in the primal is revealed. Furthermore, using this framework, we are able to introduce some novel types of constraints, allowing to impose bounds on the policy's action density or on costs associated with transitions between consecutive states and actions. From the adjusted primal-dual optimization problems, a practical algorithm is derived that supports various combinations of policy constraints that are automatically handled throughout training using trainable reward modifications. The proposed $\texttt{DualCRL}$ method is examined in more detail and evaluated under different (combinations of) constraints on two interpretable environments. The results highlight the efficacy of the method, which ultimately provides the designer of such systems with a versatile toolbox of possible policy constraints.

arXiv:2404.16399

Offline Reinforcement Learning with Behavioral Supervisor Tuning

Offline reinforcement learning (RL) algorithms are applied to learn performant, well-generalizing policies when provided with a static dataset of interactions. Many recent approaches to offline RL have seen substantial success, but with one key caveat: they demand substantial per-dataset hyperparameter tuning to achieve reported performance, which requires policy rollouts in the environment to evaluate; this can rapidly become cumbersome. Furthermore, substantial tuning requirements can hamper the adoption of these algorithms in practical domains. In this paper, we present TD3 with Behavioral Supervisor Tuning (TD3-BST), an algorithm that trains an uncertainty model and uses it to guide the policy to select actions within the dataset support. TD3-BST can learn more effective policies from offline datasets compared to previous methods and achieves the best performance across challenging benchmarks without requiring per-dataset tuning.

arXiv:2404.15617

DPO: A Differential and Pointwise Control Approach to Reinforcement Learning

Reinforcement learning (RL) in continuous state-action spaces remains challenging in scientific computing due to poor sample efficiency and lack of pathwise physical consistency. We introduce Differential Reinforcement Learning (Differential RL), a novel framework that reformulates RL from a continuous-time control perspective via a differential dual formulation. This induces a Hamiltonian structure that embeds physics priors and ensures consistent trajectories without requiring explicit constraints. To implement Differential RL, we develop Differential Policy Optimization (DPO), a pointwise, stage-wise algorithm that refines local movement operators along the trajectory for improved sample efficiency and dynamic alignment. We establish pointwise convergence guarantees, a property not available in standard RL, and derive a competitive theoretical regret bound of $O(K^{5/6})$. Empirically, DPO outperforms standard RL baselines on representative scientific computing tasks, including surface modeling, grid control, and molecular dynamics, under low-data and physics-constrained conditions.

arXiv:2404.15597

GRSN: Gated Recurrent Spiking Neurons for POMDPs and MARL

Spiking neural networks (SNNs) are widely applied in various fields due to their energy-efficient and fast-inference capabilities. Applying SNNs to reinforcement learning (RL) can significantly reduce the computational resource requirements for agents and improve the algorithm's performance under resource-constrained conditions. However, in current spiking reinforcement learning (SRL) algorithms, the simulation results of multiple time steps can only correspond to a single-step decision in RL. This is quite different from the real temporal dynamics in the brain and also fails to fully exploit the capacity of SNNs to process temporal data. In order to address this temporal mismatch issue and further take advantage of the inherent temporal dynamics of spiking neurons, we propose a novel temporal alignment paradigm (TAP) that leverages the single-step update of spiking neurons to accumulate historical state information in RL and introduces gated units to enhance the memory capacity of spiking neurons. Experimental results show that our method can solve partially observable Markov decision processes (POMDPs) and multi-agent cooperation problems with similar performance as recurrent neural networks (RNNs) but with about 50% power consumption.

arXiv:2404.15417

The Power of Resets in Online Reinforcement Learning

Simulators are a pervasive tool in reinforcement learning, but most existing algorithms cannot efficiently exploit simulator access -- particularly in high-dimensional domains that require general function approximation. We explore the power of simulators through online reinforcement learning with {local simulator access} (or, local planning), an RL protocol where the agent is allowed to reset to previously observed states and follow their dynamics during training. We use local simulator access to unlock new statistical guarantees that were previously out of reach: - We show that MDPs with low coverability (Xie et al. 2023) -- a general structural condition that subsumes Block MDPs and Low-Rank MDPs -- can be learned in a sample-efficient fashion with only $Q^{\star}$-realizability (realizability of the optimal state-value function); existing online RL algorithms require significantly stronger representation conditions. - As a consequence, we show that the notorious Exogenous Block MDP problem (Efroni et al. 2022) is tractable under local simulator access. The results above are achieved through a computationally inefficient algorithm. We complement them with a more computationally efficient algorithm, RVFS (Recursive Value Function Search), which achieves provable sample complexity guarantees under a strengthened statistical assumption known as pushforward coverability. RVFS can be viewed as a principled, provable counterpart to a successful empirical paradigm that combines recursive search (e.g., MCTS) with value function approximation.

arXiv:2404.15193

Structurally Flexible Neural Networks: Evolving the Building Blocks for General Agents

Artificial neural networks used for reinforcement learning are structurally rigid, meaning that each optimized parameter of the network is tied to its specific placement in the network structure. It also means that a network only works with pre-defined and fixed input- and output sizes. This is a consequence of having the number of optimized parameters being directly dependent on the structure of the network. Structural rigidity limits the ability to optimize parameters of policies across multiple environments that do not share input and output spaces. Here, we evolve a set of neurons and plastic synapses each represented by a gated recurrent unit (GRU). During optimization, the parameters of these fundamental units of a neural network are optimized in different random structural configurations. Earlier work has shown that parameter sharing between units is important for making structurally flexible neurons We show that it is possible to optimize a set of distinct neuron- and synapse types allowing for a mitigation of the symmetry dilemma. We demonstrate this by optimizing a single set of neurons and synapses to solve multiple reinforcement learning control tasks simultaneously.

arXiv:2404.14961

Cache-Aware Reinforcement Learning in Large-Scale Recommender Systems

Modern large-scale recommender systems are built upon computation-intensive infrastructure and usually suffer from a huge difference in traffic between peak and off-peak periods. In peak periods, it is challenging to perform real-time computation for each request due to the limited budget of computational resources. The recommendation with a cache is a solution to this problem, where a user-wise result cache is used to provide recommendations when the recommender system cannot afford a real-time computation. However, the cached recommendations are usually suboptimal compared to real-time computation, and it is challenging to determine the items in the cache for each user. In this paper, we provide a cache-aware reinforcement learning (CARL) method to jointly optimize the recommendation by real-time computation and by the cache. We formulate the problem as a Markov decision process with user states and a cache state, where the cache state represents whether the recommender system performs recommendations by real-time computation or by the cache. The computational load of the recommender system determines the cache state. We perform reinforcement learning based on such a model to improve user engagement over multiple requests. Moreover, we show that the cache will introduce a challenge called critic dependency, which deteriorates the performance of reinforcement learning. To tackle this challenge, we propose an eigenfunction learning (EL) method to learn independent critics for CARL. Experiments show that CARL can significantly improve the users' engagement when considering the result cache. CARL has been fully launched in Kwai app, serving over 100 million users.

arXiv:2404.14909

MultiSTOP: Solving Functional Equations with Reinforcement Learning

We develop MultiSTOP, a Reinforcement Learning framework for solving functional equations in physics. This new methodology produces actual numerical solutions instead of bounds on them. We extend the original BootSTOP algorithm by adding multiple constraints derived from domain-specific knowledge, even in integral form, to improve the accuracy of the solution. We investigate a particular equation in a one-dimensional Conformal Field Theory.

arXiv:2404.15059

Using deep reinforcement learning to promote sustainable human behaviour on a common pool resource problem

A canonical social dilemma arises when finite resources are allocated to a group of people, who can choose to either reciprocate with interest, or keep the proceeds for themselves. What resource allocation mechanisms will encourage levels of reciprocation that sustain the commons? Here, in an iterated multiplayer trust game, we use deep reinforcement learning (RL) to design an allocation mechanism that endogenously promotes sustainable contributions from human participants to a common pool resource. We first trained neural networks to behave like human players, creating a stimulated economy that allowed us to study how different mechanisms influenced the dynamics of receipt and reciprocation. We then used RL to train a social planner to maximise aggregate return to players. The social planner discovered a redistributive policy that led to a large surplus and an inclusive economy, in which players made roughly equal gains. The RL agent increased human surplus over baseline mechanisms based on unrestricted welfare or conditional cooperation, by conditioning its generosity on available resources and temporarily sanctioning defectors by allocating fewer resources to them. Examining the AI policy allowed us to develop an explainable mechanism that performed similarly and was more popular among players. Deep reinforcement learning can be used to discover mechanisms that promote sustainable human behaviour.

arXiv:2404.14763

Evolutionary Reinforcement Learning via Cooperative Coevolution

Recently, evolutionary reinforcement learning has obtained much attention in various domains. Maintaining a population of actors, evolutionary reinforcement learning utilises the collected experiences to improve the behaviour policy through efficient exploration. However, the poor scalability of genetic operators limits the efficiency of optimising high-dimensional neural networks.To address this issue, this paper proposes a novel cooperative coevolutionary reinforcement learning (CoERL) algorithm. Inspired by cooperative coevolution, CoERL periodically and adaptively decomposes the policy optimisation problem into multiple subproblems and evolves a population of neural networks for each of the subproblems. Instead of using genetic operators, CoERL directly searches for partial gradients to update the policy. Updating policy with partial gradients maintains consistency between the behaviour spaces of parents and offspring across generations.The experiences collected by the population are then used to improve the entire policy, which enhances the sampling efficiency.Experiments on six benchmark locomotion tasks demonstrate that CoERL outperforms seven state-of-the-art algorithms and baselines.Ablation study verifies the unique contribution of CoERL's core ingredients.

arXiv:2404.13316

On the stability of Lipschitz continuous control problems and its application to reinforcement learning

We address the crucial yet underexplored stability properties of the Hamilton--Jacobi--Bellman (HJB) equation in model-free reinforcement learning contexts, specifically for Lipschitz continuous optimal control problems. We bridge the gap between Lipschitz continuous optimal control problems and classical optimal control problems in the viscosity solutions framework, offering new insights into the stability of the value function of Lipschitz continuous optimal control problems. By introducing structural assumptions on the dynamics and reward functions, we further study the rate of convergence of value functions. Moreover, we introduce a generalized framework for Lipschitz continuous control problems that incorporates the original problem and leverage it to propose a new HJB-based reinforcement learning algorithm. The stability properties and performance of the proposed method are tested with well-known benchmark examples in comparison with existing approaches.

arXiv:2404.13423공식 코드 · Papers with Code

PIPER: Primitive-Informed Preference-based Hierarchical Reinforcement Learning via Hindsight Relabeling

In this work, we introduce PIPER: Primitive-Informed Preference-based Hierarchical reinforcement learning via Hindsight Relabeling, a novel approach that leverages preference-based learning to learn a reward model, and subsequently uses this reward model to relabel higher-level replay buffers. Since this reward is unaffected by lower primitive behavior, our relabeling-based approach is able to mitigate non-stationarity, which is common in existing hierarchical approaches, and demonstrates impressive performance across a range of challenging sparse-reward tasks. Since obtaining human feedback is typically impractical, we propose to replace the human-in-the-loop approach with our primitive-in-the-loop approach, which generates feedback using sparse rewards provided by the environment. Moreover, in order to prevent infeasible subgoal prediction and avoid degenerate solutions, we propose primitive-informed regularization that conditions higher-level policies to generate feasible subgoals for lower-level policies. We perform extensive experiments to show that PIPER mitigates non-stationarity in hierarchical reinforcement learning and achieves greater than 50$\%$ success rates in challenging, sparse-reward robotic environments, where most other baselines fail to achieve any significant progress.

arXiv:2404.14077공식 코드 · Papers with Code

Research on Robot Path Planning Based on Reinforcement Learning

This project has conducted research on robot path planning based on Visual SLAM. The main work of this project is as follows: (1) Construction of Visual SLAM system. Research has been conducted on the basic architecture of Visual SLAM. A Visual SLAM system is developed based on ORB-SLAM3 system, which can conduct dense point cloud mapping. (2) The map suitable for two-dimensional path planning is obtained through map conversion. This part converts the dense point cloud map obtained by Visual SLAM system into an octomap and then performs projection transformation to the grid map. The map conversion converts the dense point cloud map containing a large amount of redundant map information into an extremely lightweight grid map suitable for path planning. (3) Research on path planning algorithm based on reinforcement learning. This project has conducted experimental comparisons between the Q-learning algorithm, the DQN algorithm, and the SARSA algorithm, and found that DQN is the algorithm with the fastest convergence and best performance in high-dimensional complex environments. This project has conducted experimental verification of the Visual SLAM system in a simulation environment. The experimental results obtained based on open-source dataset and self-made dataset prove the feasibility and effectiveness of the designed Visual SLAM system. At the same time, this project has also conducted comparative experiments on the three reinforcement learning algorithms under the same experimental condition to obtain the optimal algorithm under the experimental condition.

arXiv:2404.14064공식 코드 · Papers with Code

Multi-view Disentanglement for Reinforcement Learning with Multiple Cameras

The performance of image-based Reinforcement Learning (RL) agents can vary depending on the position of the camera used to capture the images. Training on multiple cameras simultaneously, including a first-person egocentric camera, can leverage information from different camera perspectives to improve the performance of RL. However, hardware constraints may limit the availability of multiple cameras in real-world deployment. Additionally, cameras may become damaged in the real-world preventing access to all cameras that were used during training. To overcome these hardware constraints, we propose Multi-View Disentanglement (MVD), which uses multiple cameras to learn a policy that is robust to a reduction in the number of cameras to generalise to any single camera from the training set. Our approach is a self-supervised auxiliary task for RL that learns a disentangled representation from multiple cameras, with a shared representation that is aligned across all cameras to allow generalisation to a single camera, and a private representation that is camera-specific. We show experimentally that an RL agent trained on a single third-person camera is unable to learn an optimal policy in many control tasks; but, our approach, benefiting from multiple cameras during training, is able to solve the task using only the same single third-person camera.

arXiv:2404.13061

FPGA Divide-and-Conquer Placement using Deep Reinforcement Learning

This paper introduces the problem of learning to place logic blocks in Field-Programmable Gate Arrays (FPGAs) and a learning-based method. In contrast to previous search-based placement algorithms, we instead employ Reinforcement Learning (RL) with the goal of minimizing wirelength. In addition to our preliminary learning results, we also evaluated a novel decomposition to address the nature of large search space when placing many blocks on a chipboard. Empirical experiments evaluate the effectiveness of the learning and decomposition paradigms on FPGA placement tasks.