Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor

Abstract

Model-free deep reinforcement learning (RL) algorithms have been demonstrated on a range of challenging decision making and control tasks. However, these methods typically suffer from two major challenges: very high sample complexity and brittle convergence properties, which necessitate meticulous hyperparameter tuning. Both of these challenges severely limit the applicability of such methods to complex, real-world domains. In this paper, we propose soft actor-critic, an off-policy actor-critic deep RL algorithm based on the maximum entropy reinforcement learning framework. In this framework, the actor aims to maximize expected reward while also maximizing entropy. That is, to succeed at the task while acting as randomly as possible. Prior deep RL methods based on this framework have been formulated as Q-learning methods. By combining off-policy updates with a stable stochastic actor-critic formulation, our method achieves state-of-the-art performance on a range of continuous control benchmark tasks, outperforming prior on-policy and off-policy methods. Furthermore, we demonstrate that, in contrast to other off-policy algorithms, our approach is very stable, achieving very similar performance across different random seeds.

arXiv commentICML 2018 Videos: sites.google.com/view/soft-actor-critic Code: github.com/haarnoja/sac

References in corpus (10)

• Continuous control with deep reinforcement learning
• Playing Atari with Deep Reinforcement Learning
• Benchmarking Deep Reinforcement Learning for Continuous Control
• Learning Continuous Control Policies by Stochastic Value Gradients
• Bridging the Gap Between Value and Policy Based Reinforcement Learning
• Equivalence Between Policy Gradients and Soft Q-Learning
• Data-efficient Deep Reinforcement Learning for Dexterous Manipulation
• Q-Prop: Sample-Efficient Policy Gradient with An Off-Policy Critic
• Combining policy gradient and Q-learning
• Trust-PCL: An Off-Policy Trust Region Method for Continuous Control

Cited by in corpus (798)

• Addressing Function Approximation Error in Actor-Critic Methods
• Soft Actor-Critic Algorithms and Applications
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• A Survey and Critique of Multiagent Deep Reinforcement Learning
• How to Train Your Robot with Deep Reinforcement Learning; Lessons We've Learned
• Unity: A General Platform for Intelligent Agents
• A Review of Deep Reinforcement Learning for Smart Building Energy Management
• Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges
• Multi-Task Learning with Deep Neural Networks: A Survey
• Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction
• Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
• Distributional Soft Actor-Critic: Off-Policy Reinforcement Learning for Addressing Value Estimation Errors
• D4RL: Datasets for Deep Data-Driven Reinforcement Learning
• Out-of-Distribution Generalization via Risk Extrapolation (REx)
• Behavior Regularized Offline Reinforcement Learning
• MOPO: Model-based Offline Policy Optimization
• Benchmarking Model-Based Reinforcement Learning
• Deep Reinforcement Learning Attitude Control of Fixed-Wing UAVs Using Proximal Policy Optimization
• Federated Reinforcement Learning: Techniques, Applications, and Open Challenges
• Text2Motion: From Natural Language Instructions to Feasible Plans
• Soft Actor-Critic for Discrete Action Settings
• Resource Management in Wireless Networks via Multi-Agent Deep Reinforcement Learning
• Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
• A Minimalist Approach to Offline Reinforcement Learning
• Simple random search provides a competitive approach to reinforcement learning
• Learning Force Control for Contact-rich Manipulation Tasks with Rigid Position-controlled Robots
• MOReL : Model-Based Offline Reinforcement Learning
• Transfer Learning in Deep Reinforcement Learning: A Survey
• Autonomous Unmanned Aerial Vehicle Navigation using Reinforcement Learning: A Systematic Review
• Game-Theoretic Multiagent Reinforcement Learning
• RLOC: Terrain-Aware Legged Locomotion using Reinforcement Learning and Optimal Control
• Dream to Control: Learning Behaviors by Latent Imagination
• Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees
• Deep Reinforcement Learning with Shallow Controllers: An Experimental Application to PID Tuning
• Prioritized Experience-based Reinforcement Learning with Human Guidance for Autonomous Driving
• Tianshou: a Highly Modularized Deep Reinforcement Learning Library
• When to Trust Your Model: Model-Based Policy Optimization
• Identifying optimal cycles in quantum thermal machines with reinforcement-learning
• Reinforcement Learning Algorithms: An Overview and Classification
• What Matters In On-Policy Reinforcement Learning? A Large-Scale Empirical Study
• Gradient Surgery for Multi-Task Learning
• FACMAC: Factored Multi-Agent Centralised Policy Gradients
• A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms
• An Optimistic Perspective on Offline Reinforcement Learning
• Deep Reinforcement Learning for Black-Box Testing of Android Apps
• Energy Management Based on Multi-Agent Deep Reinforcement Learning for A Multi-Energy Industrial Park
• Learning to Walk in the Real World with Minimal Human Effort
• Generative Adversarial Imitation from Observation
• Efficient Exploration via State Marginal Matching
• Horizon: Facebook's Open Source Applied Reinforcement Learning Platform
• Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning
• Reinforcement Learning for Many-Body Ground-State Preparation Inspired by Counterdiabatic Driving
• On the Global Convergence Rates of Softmax Policy Gradient Methods
• AlgaeDICE: Policy Gradient from Arbitrary Experience
• Data-Efficient Deep Reinforcement Learning for Attitude Control of Fixed-Wing UAVs: Field Experiments
• Fusion of Model-free Reinforcement Learning with Microgrid Control: Review and Vision
• Variational Option Discovery Algorithms
• Learning Invariant Representations for Reinforcement Learning without Reconstruction
• GEP-PG: Decoupling Exploration and Exploitation in Deep Reinforcement Learning Algorithms
• Exploring Model-based Planning with Policy Networks
• Active Learning in Robotics: A Review of Control Principles
• DRLinFluids -- An open-source python platform of coupling Deep Reinforcement Learning and OpenFOAM
• Autonomous Navigation of an Ultrasound Probe Towards Standard Scan Planes with Deep Reinforcement Learning
• Dynamics-Aware Unsupervised Discovery of Skills
• AWAC: Accelerating Online Reinforcement Learning with Offline Datasets
• t-Soft Update of Target Network for Deep Reinforcement Learning
• Robust High-speed Running for Quadruped Robots via Deep Reinforcement Learning
• Adaptive dynamic programming for nonaffine nonlinear optimal control problem with state constraints
• Latent Space Policies for Hierarchical Reinforcement Learning
• Autonomous Navigation for Robot-assisted Intraluminal and Endovascular Procedures: A Systematic Review
• SoftGym: Benchmarking Deep Reinforcement Learning for Deformable Object Manipulation
• Deep Dynamics Models for Learning Dexterous Manipulation
• Decoupling Representation Learning from Reinforcement Learning
• Precise atom manipulation through deep reinforcement learning
• Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model
• iGibson 2.0: Object-Centric Simulation for Robot Learning of Everyday Household Tasks
• CityLearn v2: Energy-flexible, resilient, occupant-centric, and carbon-aware management of grid-interactive communities
• Queueing Network Controls via Deep Reinforcement Learning
• Online Service Migration in Mobile Edge with Incomplete System Information: A Deep Recurrent Actor-Critic Learning Approach
• Multi-Task Reinforcement Learning with Soft Modularization
• Meta reinforcement learning as task inference
• Reinforcement Learning through Active Inference
• A Self-adaptive SAC-PID Control Approach based on Reinforcement Learning for Mobile Robots
• A Perspective on Deep Learning for Molecular Modeling and Simulations
• Neurosymbolic Reinforcement Learning and Planning: A Survey
• Avoiding Catastrophe: Active Dendrites Enable Multi-Task Learning in Dynamic Environments
• An experimental evaluation of Deep Reinforcement Learning algorithms for HVAC control
• Residual Learning from Demonstration: Adapting DMPs for Contact-rich Manipulation
• AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search
• Unsupervised Meta-Learning for Reinforcement Learning
• Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics
• Automated deep reinforcement learning for real-time scheduling strategy of multi-energy system integrated with post-carbon and direct-air carbon captured system
• COMBO: Conservative Offline Model-Based Policy Optimization
• Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning
• First Order Constrained Optimization in Policy Space
• Multi-Task Fusion via Reinforcement Learning for Long-Term User Satisfaction in Recommender Systems
• An information-theoretic perspective on intrinsic motivation in reinforcement learning: a survey
• SQIL: Imitation Learning via Reinforcement Learning with Sparse Rewards
• rlpyt: A Research Code Base for Deep Reinforcement Learning in PyTorch
• Learning to swim in potential flow
• Phasic Policy Gradient
• A reinforcement learning approach to rare trajectory sampling
• Deep symbolic regression: Recovering mathematical expressions from data via risk-seeking policy gradients
• Uncertainty-Based Offline Reinforcement Learning with Diversified Q-Ensemble
• Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning
• Softmax Deep Double Deterministic Policy Gradients
• Learning Implicit Credit Assignment for Cooperative Multi-Agent Reinforcement Learning
• The Free Energy Principle for Perception and Action: A Deep Learning Perspective
• SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning
• SACHA: Soft Actor-Critic with Heuristic-Based Attention for Partially Observable Multi-Agent Path Finding
• Reinforcement Learning Applications
• Longitudinal Dynamic versus Kinematic Models for Car-Following Control Using Deep Reinforcement Learning
• Modern Machine Learning Tools for Monitoring and Control of Industrial Processes: A Survey
• A Bayesian Framework for Digital Twin-Based Control, Monitoring, and Data Collection in Wireless Systems
• Gym-ANM: Reinforcement Learning Environments for Active Network Management Tasks in Electricity Distribution Systems
• InfoBot: Transfer and Exploration via the Information Bottleneck
• Measuring the Reliability of Reinforcement Learning Algorithms
• Relative Entropy Regularized Policy Iteration
• Deep Reinforcement Learning at the Edge of the Statistical Precipice
• Model-Augmented Actor-Critic: Backpropagating through Paths
• Boosting Soft Actor-Critic: Emphasizing Recent Experience without Forgetting the Past
• Generative Adversarial Self-Imitation Learning
• Fast Global Convergence of Natural Policy Gradient Methods with Entropy Regularization
• Discrete and Continuous Action Representation for Practical RL in Video Games
• V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control
• Neural Ordinary Differential Equation Control of Dynamics on Graphs
• Primal Wasserstein Imitation Learning
• Safety and Liveness Guarantees through Reach-Avoid Reinforcement Learning
• Deep Implicit Coordination Graphs for Multi-agent Reinforcement Learning
• Variational Inference MPC for Bayesian Model-based Reinforcement Learning
• Stabilizing Deep Q-Learning with ConvNets and Vision Transformers under Data Augmentation
• Invariant Causal Prediction for Block MDPs
• A Comprehensive Overview and Survey of Recent Advances in Meta-Learning
• Inclined Quadrotor Landing using Deep Reinforcement Learning
• Reinforcement Learning for Robotic Manipulation using Simulated Locomotion Demonstrations
• Deep Reinforcement Learning with Population-Coded Spiking Neural Network for Continuous Control
• Reinforcement learning of rare diffusive dynamics
• A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
• Robust Biped Locomotion Using Deep Reinforcement Learning on Top of an Analytical Control Approach
• DisCor: Corrective Feedback in Reinforcement Learning via Distribution Correction
• APPLE: Adaptive Planner Parameter Learning from Evaluative Feedback
• TD-Regularized Actor-Critic Methods
• Human-Centric Resource Allocation for the Metaverse With Multiaccess Edge Computing
• Adversarial Distributional Training for Robust Deep Learning
• Learn to Navigate Maplessly with Varied LiDAR Configurations: A Support Point-Based Approach
• Coordinated Exploration via Intrinsic Rewards for Multi-Agent Reinforcement Learning
• One model Packs Thousands of Items with Recurrent Conditional Query Learning
• If MaxEnt RL is the Answer, What is the Question?
• The Forecast Trap
• Conservative Safety Critics for Exploration
• BAIL: Best-Action Imitation Learning for Batch Deep Reinforcement Learning
• Interactive Differentiable Simulation
• Truly Proximal Policy Optimization
• Robust Quadruped Jumping via Deep Reinforcement Learning
• Safe Reinforcement Learning Using Black-Box Reachability Analysis
• CausalWorld: A Robotic Manipulation Benchmark for Causal Structure and Transfer Learning
• Resilient Control of Networked Microgrids using Vertical Federated Reinforcement Learning: Designs and Real-Time Test-Bed Validations
• FreeKD: Free-direction Knowledge Distillation for Graph Neural Networks
• Rewriting History with Inverse RL: Hindsight Inference for Policy Improvement
• Model-free Deep Reinforcement Learning for Urban Autonomous Driving
• VIREL: A Variational Inference Framework for Reinforcement Learning
• Trajectory Planning with Deep Reinforcement Learning in High-Level Action Spaces
• N${}^2$M${}^2$: Learning Navigation for Arbitrary Mobile Manipulation Motions in Unseen and Dynamic Environments
• Maximum Entropy RL (Provably) Solves Some Robust RL Problems
• Learning Robotic Navigation from Experience: Principles, Methods, and Recent Results
• Data Efficient Reinforcement Learning for Legged Robots
• Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation
• Learning Shape Control of Elastoplastic Deformable Linear Objects
• EpidemiOptim: A Toolbox for the Optimization of Control Policies in Epidemiological Models
• Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learning
• Model-Based Safe Reinforcement Learning with Time-Varying State and Control Constraints: An Application to Intelligent Vehicles
• IPAPRec: A promising tool for learning high-performance mapless navigation skills with deep reinforcement learning
• A Divergence Minimization Perspective on Imitation Learning Methods
• Interpretable End-to-end Urban Autonomous Driving with Latent Deep Reinforcement Learning
• When Cyber-Physical Systems Meet AI: A Benchmark, an Evaluation, and a Way Forward
• Hierarchical Reinforcement Learning By Discovering Intrinsic Options
• Benchmarking Actor-Critic Deep Reinforcement Learning Algorithms for Robotics Control with Action Constraints
• Model-Based Multi-Agent RL in Zero-Sum Markov Games with Near-Optimal Sample Complexity
• TriFinger: An Open-Source Robot for Learning Dexterity
• Model-based Reinforcement Learning for Semi-Markov Decision Processes with Neural ODEs
• Mirror Descent Policy Optimization
• Recall Traces: Backtracking Models for Efficient Reinforcement Learning
• Action and Perception as Divergence Minimization
• Forward and inverse reinforcement learning sharing network weights and hyperparameters
• Benchmarks for Deep Off-Policy Evaluation
• Sample-efficient Reinforcement Learning Representation Learning with Curiosity Contrastive Forward Dynamics Model
• Generative Slate Recommendation with Reinforcement Learning
• How to pick the domain randomization parameters for sim-to-real transfer of reinforcement learning policies?
• Optimistic Reinforcement Learning by Forward Kullback-Leibler Divergence Optimization
• On-Policy Robot Imitation Learning from a Converging Supervisor
• Deep Reinforcement Learning for Human-Like Driving Policies in Collision Avoidance Tasks of Self-Driving Cars
• DoorGym: A Scalable Door Opening Environment And Baseline Agent
• Towards Practical Multi-Object Manipulation using Relational Reinforcement Learning
• Energy-Efficient Control Adaptation with Safety Guarantees for Learning-Enabled Cyber-Physical Systems
• A Centralised Soft Actor Critic Deep Reinforcement Learning Approach to District Demand Side Management through CityLearn
• Infusing model predictive control into meta-reinforcement learning for mobile robots in dynamic environments
• State Entropy Maximization with Random Encoders for Efficient Exploration
• Optimal Stroke Learning with Policy Gradient Approach for Robotic Table Tennis
• Unsupervised Social Event Detection via Hybrid Graph Contrastive Learning and Reinforced Incremental Clustering
• Improving robot navigation in crowded environments using intrinsic rewards
• Exploration versus exploitation in reinforcement learning: a stochastic control approach
• Generalizable Episodic Memory for Deep Reinforcement Learning
• Continual World: A Robotic Benchmark For Continual Reinforcement Learning
• Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environments
• Accelerating Quadratic Optimization with Reinforcement Learning
• Physics-informed Modularized Neural Network for Advanced Building Control by Deep Reinforcement Learning
• Recovery RL: Safe Reinforcement Learning with Learned Recovery Zones
• Boosting Trust Region Policy Optimization by Normalizing Flows Policy
• Generalization in Transfer Learning
• A0C: Alpha Zero in Continuous Action Space
• Evolving Reinforcement Learning Algorithms
• Online Meta-Critic Learning for Off-Policy Actor-Critic Methods
• Explanation-Aware Experience Replay in Rule-Dense Environments
• S-RL Toolbox: Environments, Datasets and Evaluation Metrics for State Representation Learning
• Dropout Q-Functions for Doubly Efficient Reinforcement Learning
• Trust the Model When It Is Confident: Masked Model-based Actor-Critic
• Learning a Decentralized Multi-arm Motion Planner
• Value Functions Factorization with Latent State Information Sharing in Decentralized Multi-Agent Policy Gradients
• Error Bounds of Imitating Policies and Environments
• Experience Replay with Likelihood-free Importance Weights
• Deep reinforcement learning for smart calibration of radio telescopes
• Generalized Hindsight for Reinforcement Learning
• Robotic Table Tennis: A Case Study into a High Speed Learning System
• Learning to Manipulate Deformable Objects without Demonstrations
• Meta-Reinforcement Learning Robust to Distributional Shift via Model Identification and Experience Relabeling
• Distributionally Robust Reinforcement Learning
• A View on Deep Reinforcement Learning in System Optimization
• Meta-SAC: Auto-tune the Entropy Temperature of Soft Actor-Critic via Metagradient
• State Alignment-based Imitation Learning
• Integrating Contrastive Learning with Dynamic Models for Reinforcement Learning from Images
• Dynamics Generalization via Information Bottleneck in Deep Reinforcement Learning
• Efficient Off-Policy Safe Reinforcement Learning Using Trust Region Conditional Value at Risk
• Delay-Aware Multi-Agent Reinforcement Learning for Cooperative and Competitive Environments
• PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics
• Regularization Matters in Policy Optimization
• Dynamics-aware Embeddings
• AC-Teach: A Bayesian Actor-Critic Method for Policy Learning with an Ensemble of Suboptimal Teachers
• A physics-informed reinforcement learning approach for the interfacial area transport in two-phase flow
• Strategies for Using Proximal Policy Optimization in Mobile Puzzle Games
• Inverse Rational Control with Partially Observable Continuous Nonlinear Dynamics
• Deep Reinforcement Learning for Time-Critical Wilderness Search And Rescue Using Drones
• Combining Q-Learning and Search with Amortized Value Estimates
• Offline Reinforcement Learning from Images with Latent Space Models
• A Unified Bellman Optimality Principle Combining Reward Maximization and Empowerment
• An Information-Theoretic Optimality Principle for Deep Reinforcement Learning
• Motion Planner Augmented Reinforcement Learning for Robot Manipulation in Obstructed Environments
• Hierarchical Reinforcement Learning Framework for Stochastic Spaceflight Campaign Design
• Single-Timescale Actor-Critic Provably Finds Globally Optimal Policy
• Hindsight Goal Ranking on Replay Buffer for Sparse Reward Environment
• Deep Reinforcement Learning for Contact-Rich Skills Using Compliant Movement Primitives
• OPAL: Offline Primitive Discovery for Accelerating Offline Reinforcement Learning
• Learning Dense Visual Correspondences in Simulation to Smooth and Fold Real Fabrics
• Q-Learning for Continuous Actions with Cross-Entropy Guided Policies
• Watch, Try, Learn: Meta-Learning from Demonstrations and Reward
• Can Increasing Input Dimensionality Improve Deep Reinforcement Learning?
• Improving Exploration in Soft-Actor-Critic with Normalizing Flows Policies
• Supplementing Recurrent Neural Networks with Annealing to Solve Combinatorial Optimization Problems
• Provable Benefits of Actor-Critic Methods for Offline Reinforcement Learning
• Discretizing Continuous Action Space for On-Policy Optimization
• Reinforcement Learning for Optimization of COVID-19 Mitigation policies
• Remember and Forget for Experience Replay
• Policy Evaluation Networks
• Heuristic-Guided Reinforcement Learning
• Privacy-Preserving Federated Learning for UAV-Enabled Networks: Learning-Based Joint Scheduling and Resource Management
• GIN: Graph-based Interaction-aware Constraint Policy Optimization for Autonomous Driving
• Learning Task-relevant Representations for Generalization via Characteristic Functions of Reward Sequence Distributions
• Heterogeneous Multi-Agent Reinforcement Learning for Unknown Environment Mapping
• Behavior Priors for Efficient Reinforcement Learning
• Vulnerability-Aware Poisoning Mechanism for Online RL with Unknown Dynamics
• Smoother Entropy for Active State Trajectory Estimation and Obfuscation in POMDPs
• Deep Reinforcement Learning for Autonomous Internet of Things: Model, Applications and Challenges
• On the Estimation Bias in Double Q-Learning
• Online 3D Bin Packing with Constrained Deep Reinforcement Learning
• Striving for Simplicity and Performance in Off-Policy DRL: Output Normalization and Non-Uniform Sampling
• Observation Space Matters: Benchmark and Optimization Algorithm
• Statistical Bootstrapping for Uncertainty Estimation in Off-Policy Evaluation
• CAQL: Continuous Action Q-Learning
• D2RL: Deep Dense Architectures in Reinforcement Learning
• Learning Dexterous Manipulation from Suboptimal Experts
• Learning Latent Representations to Influence Multi-Agent Interaction
• Optimal regularizations for data generation with probabilistic graphical models
• Neural Temporal-Difference and Q-Learning Provably Converge to Global Optima
• Ctrl-Z: Recovering from Instability in Reinforcement Learning
• UniCon: Universal Neural Controller For Physics-based Character Motion
• Tactical Optimism and Pessimism for Deep Reinforcement Learning
• Learning Latent Representations to Co-Adapt to Humans
• Energy-Based Imitation Learning
• A Survey of Reinforcement Learning Techniques: Strategies, Recent Development, and Future Directions
• Robust Reinforcement Learning via Adversarial training with Langevin Dynamics
• Reinforcement Learning for Personalized Dialogue Management
• Sample Efficient Reinforcement Learning via Low-Rank Matrix Estimation
• Model-Free Approach to Fair Solar PV Curtailment Using Reinforcement Learning
• Decomposed Soft Actor-Critic Method for Cooperative Multi-Agent Reinforcement Learning
• Reset-Free Lifelong Learning with Skill-Space Planning
• Fairness-aware Competitive Bidding Influence Maximization in Social Networks
• Learning to Collide: An Adaptive Safety-Critical Scenarios Generating Method
• Reward Guidance for Reinforcement Learning Tasks Based on Large Language Models: The LMGT Framework
• Soft Hindsight Experience Replay
• ROI-Constrained Bidding via Curriculum-Guided Bayesian Reinforcement Learning
• TC-Driver: Trajectory Conditioned Driving for Robust Autonomous Racing -- A Reinforcement Learning Approach
• Leveraging exploration in off-policy algorithms via normalizing flows
• Optimistic Policy Optimization with Bandit Feedback
• From Scratch to Sketch: Deep Decoupled Hierarchical Reinforcement Learning for Robotic Sketching Agent
• Low-Precision Reinforcement Learning: Running Soft Actor-Critic in Half Precision
• Behavior From the Void: Unsupervised Active Pre-Training
• Cloud-Edge Training Architecture for Sim-to-Real Deep Reinforcement Learning
• A Safe Reinforcement Learning Algorithm for Supervisory Control of Power Plants
• SpikePropamine: Differentiable Plasticity in Spiking Neural Networks
• A Tutorial on Sparse Gaussian Processes and Variational Inference
• Self-Imitation Learning via Generalized Lower Bound Q-learning
• Exploration and preference satisfaction trade-off in reward-free learning
• Offline Meta-Reinforcement Learning with Advantage Weighting
• Off-Dynamics Reinforcement Learning: Training for Transfer with Domain Classifiers
• Deep Reinforcement Learning in a Monetary Model
• Emergent Real-World Robotic Skills via Unsupervised Off-Policy Reinforcement Learning
• End-to-End Reinforcement Learning of Koopman Models for Economic Nonlinear Model Predictive Control
• Enhancing Cyber Resilience of Networked Microgrids using Vertical Federated Reinforcement Learning
• Learning Action-Transferable Policy with Action Embedding
• Adaptive and Multiple Time-scale Eligibility Traces for Online Deep Reinforcement Learning
• Training Larger Networks for Deep Reinforcement Learning
• Skip Training for Multi-Agent Reinforcement Learning Controller for Industrial Wave Energy Converters
• XIRL: Cross-embodiment Inverse Reinforcement Learning
• Divergence-Augmented Policy Optimization
• Context Meta-Reinforcement Learning via Neuromodulation
• A Memory Efficient Deep Reinforcement Learning Approach For Snake Game Autonomous Agents
• Off-policy Maximum Entropy Reinforcement Learning : Soft Actor-Critic with Advantage Weighted Mixture Policy(SAC-AWMP)
• Neural Architecture Search using Deep Neural Networks and Monte Carlo Tree Search
• Proximal Policy Optimization with Relative Pearson Divergence
• Revisiting Parameter Sharing in Multi-Agent Deep Reinforcement Learning
• Pipeline PSRO: A Scalable Approach for Finding Approximate Nash Equilibria in Large Games
• Cooperative Multi-Agent Transfer Learning with Level-Adaptive Credit Assignment
• Multi-task Batch Reinforcement Learning with Metric Learning
• MULTIPOLAR: Multi-Source Policy Aggregation for Transfer Reinforcement Learning between Diverse Environmental Dynamics
• Model-Based Opponent Modeling
• Uncertainty-aware Model-based Policy Optimization
• Disentangled Skill Embeddings for Reinforcement Learning
• Kalman meets Bellman: Improving Policy Evaluation through Value Tracking
• Learning to Collaborate in Multi-Module Recommendation via Multi-Agent Reinforcement Learning without Communication
• Batch-Constrained Reinforcement Learning for Dynamic Distribution Network Reconfiguration
• C-Learning: Learning to Achieve Goals via Recursive Classification
• ACL-QL: Adaptive Conservative Level in Q-Learning for Offline Reinforcement Learning
• Proximal Policy Optimization with Adaptive Threshold for Symmetric Relative Density Ratio
• Neural Stochastic Dual Dynamic Programming
• Optimization Landscape of Gradient Descent for Discrete-time Static Output Feedback
• Greedification Operators for Policy Optimization: Investigating Forward and Reverse KL Divergences
• Noise-Robust End-to-End Quantum Control using Deep Autoregressive Policy Networks
• One Solution is Not All You Need: Few-Shot Extrapolation via Structured MaxEnt RL
• Towards Effective Context for Meta-Reinforcement Learning: an Approach based on Contrastive Learning
• Offline Meta Learning of Exploration
• Dreaming: Model-based Reinforcement Learning by Latent Imagination without Reconstruction
• MetaCURE: Meta Reinforcement Learning with Empowerment-Driven Exploration
• Diversity Actor-Critic: Sample-Aware Entropy Regularization for Sample-Efficient Exploration
• Smooth Exploration for Robotic Reinforcement Learning
• Reinforcement Learning for Active Flow Control in Experiments
• Integrating Behavior Cloning and Reinforcement Learning for Improved Performance in Dense and Sparse Reward Environments
• Multi-Agent Actor-Critic with Hierarchical Graph Attention Network
• Sim2Real for Peg-Hole Insertion with Eye-in-Hand Camera
• OffWorld Gym: open-access physical robotics environment for real-world reinforcement learning benchmark and research
• Multimodal Safety-Critical Scenarios Generation for Decision-Making Algorithms Evaluation
• Fast and Efficient Locomotion via Learned Gait Transitions
• Generalized Second Order Value Iteration in Markov Decision Processes
• Policy Mirror Descent for Regularized Reinforcement Learning: A Generalized Framework with Linear Convergence
• Trajectory Design for UAV-Based Internet-of-Things Data Collection: A Deep Reinforcement Learning Approach
• A reinforcement learning path planning approach for range-only underwater target localization with autonomous vehicles
• Fever Basketball: A Complex, Flexible, and Asynchronized Sports Game Environment for Multi-agent Reinforcement Learning
• The Effect of Multi-step Methods on Overestimation in Deep Reinforcement Learning
• End-to-End Reinforcement Learning for Torque Based Variable Height Hopping
• Which Mutual-Information Representation Learning Objectives are Sufficient for Control?
• Safety Augmented Value Estimation from Demonstrations (SAVED): Safe Deep Model-Based RL for Sparse Cost Robotic Tasks
• Provable Fictitious Play for General Mean-Field Games
• Direct and indirect reinforcement learning
• ROLL: Visual Self-Supervised Reinforcement Learning with Object Reasoning
• $H_{\infty }$ Model-free Reinforcement Learning with Robust Stability Guarantee
• Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planning
• Value Iteration in Continuous Actions, States and Time
• Efficient reinforcement learning control for continuum robots based on Inexplicit Prior Knowledge
• Delay-Aware Model-Based Reinforcement Learning for Continuous Control
• Disentangling Adaptive Gradient Methods from Learning Rates
• Attraction-Repulsion Actor-Critic for Continuous Control Reinforcement Learning
• Doubly Robust Off-Policy Actor-Critic: Convergence and Optimality
• Improving Generalization of Reinforcement Learning with Minimax Distributional Soft Actor-Critic
• Flow Rate Control in Smart District Heating Systems Using Deep Reinforcement Learning
• Weakly Supervised Disentangled Representation for Goal-conditioned Reinforcement Learning
• Investigation of reinforcement learning for shape optimization of profile extrusion dies
• MARLIN: Soft Actor-Critic based Reinforcement Learning for Congestion Control in Real Networks
• VFunc: a Deep Generative Model for Functions
• Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling
• Accelerated Intravascular Ultrasound Imaging using Deep Reinforcement Learning
• Tonic: A Deep Reinforcement Learning Library for Fast Prototyping and Benchmarking
• Entropy-Aware Model Initialization for Effective Exploration in Deep Reinforcement Learning
• Contrastive Variational Reinforcement Learning for Complex Observations
• Stable Policy Optimization via Off-Policy Divergence Regularization
• Population-coding and Dynamic-neurons improved Spiking Actor Network for Reinforcement Learning
• SoundSpaces: Audio-Visual Navigation in 3D Environments
• Scalable Synthesis of Verified Controllers in Deep Reinforcement Learning
• Local Search for Policy Iteration in Continuous Control
• EMaQ: Expected-Max Q-Learning Operator for Simple Yet Effective Offline and Online RL
• Measuring and Harnessing Transference in Multi-Task Learning
• A Hybrid Stochastic Policy Gradient Algorithm for Reinforcement Learning
• A Workflow for Offline Model-Free Robotic Reinforcement Learning
• Accelerating Reinforcement Learning Agent with EEG-based Implicit Human Feedback
• Continuous Doubly Constrained Batch Reinforcement Learning
• Hierarchical Decomposition of Nonlinear Dynamics and Control for System Identification and Policy Distillation
• Efficient and Robust Reinforcement Learning with Uncertainty-based Value Expansion
• Plan2Vec: Unsupervised Representation Learning by Latent Plans
• Hard instance learning for quantum adiabatic prime factorization
• Diversity-Enriched Option-Critic
• Bridging Imagination and Reality for Model-Based Deep Reinforcement Learning
• Information Theoretic Model Predictive Q-Learning
• Migration of self-propelling agent in a turbulent environment with minimal energy consumption
• Learning Vision-based Robotic Manipulation Tasks Sequentially in Offline Reinforcement Learning Settings
• Reinforcement Learning by Guided Safe Exploration
• Implicit Policy for Reinforcement Learning
• Learning Practically Feasible Policies for Online 3D Bin Packing
• Learning Cooperative Multi-Agent Policies with Partial Reward Decoupling
• Model-based Policy Optimization with Unsupervised Model Adaptation
• A Regularized Opponent Model with Maximum Entropy Objective
• SLM Lab: A Comprehensive Benchmark and Modular Software Framework for Reproducible Deep Reinforcement Learning
• Reverb: A Framework For Experience Replay
• Low Emission Building Control with Zero-Shot Reinforcement Learning
• Implicit Under-Parameterization Inhibits Data-Efficient Deep Reinforcement Learning
• MUSBO: Model-based Uncertainty Regularized and Sample Efficient Batch Optimization for Deployment Constrained Reinforcement Learning
• Jointly-Learned State-Action Embedding for Efficient Reinforcement Learning
• Locally Persistent Exploration in Continuous Control Tasks with Sparse Rewards
• Hindsight Expectation Maximization for Goal-conditioned Reinforcement Learning
• Learning Markov State Abstractions for Deep Reinforcement Learning
• Domain Adversarial Reinforcement Learning
• Maximum Reward Formulation In Reinforcement Learning
• Towards Off-Policy Reinforcement Learning for Ranking Policies with Human Feedback
• Online Hyper-parameter Tuning in Off-policy Learning via Evolutionary Strategies
• A Maximum Mutual Information Framework for Multi-Agent Reinforcement Learning
• Deep RL With Information Constrained Policies: Generalization in Continuous Control
• Generalized Off-Policy Actor-Critic
• Provably Convergent Two-Timescale Off-Policy Actor-Critic with Function Approximation
• MRAC-RL: A Framework for On-Line Policy Adaptation Under Parametric Model Uncertainty
• Task-Agnostic Morphology Evolution
• Breaking the Deadly Triad with a Target Network
• Safe Autonomous Racing via Approximate Reachability on Ego-vision
• Advances in Preference-based Reinforcement Learning: A Review
• Data-efficient Co-Adaptation of Morphology and Behaviour with Deep Reinforcement Learning
• Constrained Model-Free Reinforcement Learning for Process Optimization
• Deep Reinforcement Learning with Feedback-based Exploration
• Learning to Engage with Interactive Systems: A Field Study on Deep Reinforcement Learning in a Public Museum
• Only Relevant Information Matters: Filtering Out Noisy Samples to Boost RL
• Dealing with Sparse Rewards Using Graph Neural Networks
• Teach Biped Robots to Walk via Gait Principles and Reinforcement Learning with Adversarial Critics
• Learning Object-conditioned Exploration using Distributed Soft Actor Critic
• Is the Policy Gradient a Gradient?
• Visual Adversarial Imitation Learning using Variational Models
• Adversarial Skill Chaining for Long-Horizon Robot Manipulation via Terminal State Regularization
• Weakly-Supervised Reinforcement Learning for Controllable Behavior
• Distributional Actor-Critic Ensemble for Uncertainty-Aware Continuous Control
• A Deeper Look at Discounting Mismatch in Actor-Critic Algorithms
• Improving Post-Processing of Audio Event Detectors Using Reinforcement Learning
• Non-local Policy Optimization via Diversity-regularized Collaborative Exploration
• Modelling Bounded Rationality in Multi-Agent Interactions by Generalized Recursive Reasoning
• Learning to Navigate in Turbulent Flows with Aerial Robot Swarms: A Cooperative Deep Reinforcement Learning Approach
• A Max-Min Entropy Framework for Reinforcement Learning
• Reinforcement Learning with Videos: Combining Offline Observations with Interaction
• GRAC: Self-Guided and Self-Regularized Actor-Critic
• Model Imitation for Model-Based Reinforcement Learning
• Cooperative Highway Work Zone Merge Control based on Reinforcement Learning in A Connected and Automated Environment
• Seeing All the Angles: Learning Multiview Manipulation Policies for Contact-Rich Tasks from Demonstrations
• Learning Robust State Abstractions for Hidden-Parameter Block MDPs
• Reinforcement Learning Approach to Vibration Compensation for Dynamic Feed Drive Systems
• Multi-Task Multi-Behavior MAP-Elites
• Mean-Variance Policy Iteration for Risk-Averse Reinforcement Learning
• Meta Automatic Curriculum Learning
• S4RL: Surprisingly Simple Self-Supervision for Offline Reinforcement Learning
• Generalized Decision Transformer for Offline Hindsight Information Matching
• Offline Reinforcement Learning with Pseudometric Learning
• Off-Policy Actor-Critic with Shared Experience Replay
• Meta-Model-Based Meta-Policy Optimization
• Self-organization of action hierarchy and compositionality by reinforcement learning with recurrent neural networks
• Deep Reinforcement Learning for Process Synthesis
• Caching Transient Content for IoT Sensing: Multi-Agent Soft Actor-Critic
• Learning from Ambiguous Demonstrations with Self-Explanation Guided Reinforcement Learning
• Human-in-the-Loop Methods for Data-Driven and Reinforcement Learning Systems
• A Pragmatic Look at Deep Imitation Learning
• Sequoia: A Software Framework to Unify Continual Learning Research
• Ensuring Monotonic Policy Improvement in Entropy-regularized Value-based Reinforcement Learning
• Meta-Reinforcement Learning in Broad and Non-Parametric Environments
• Conformal Bootstrap with Reinforcement Learning
• Band-limited Soft Actor Critic Model
• A Dual Approach to Constrained Markov Decision Processes with Entropy Regularization
• Signs of Language: Embodied Sign Language Fingerspelling Acquisition from Demonstrations for Human-Robot Interaction
• RITA: Boost Driving Simulators with Realistic Interactive Traffic Flow
• Learning to Manipulate Object Collections Using Grounded State Representations
• Model Based Planning with Energy Based Models
• Soft Policy Gradient Method for Maximum Entropy Deep Reinforcement Learning
• Replacing Rewards with Examples: Example-Based Policy Search via Recursive Classification
• VILD: Variational Imitation Learning with Diverse-quality Demonstrations
• AMBER: Adaptive Multi-Batch Experience Replay for Continuous Action Control
• HiER: Highlight Experience Replay for Boosting Off-Policy Reinforcement Learning Agents
• Learning Long-Term Reward Redistribution via Randomized Return Decomposition
• Implicit Generative Modeling for Efficient Exploration
• Continuous-Time Model-Based Reinforcement Learning
• Adaptive Torque Control of Exoskeletons under Spasticity Conditions via Reinforcement Learning
• Regularized Behavior Value Estimation
• A Learning Framework for High Precision Industrial Assembly
• Design of Restricted Normalizing Flow towards Arbitrary Stochastic Policy with Computational Efficiency
• Quinoa: a Q-function You Infer Normalized Over Actions
• Reward Tweaking: Maximizing the Total Reward While Planning for Short Horizons
• Learning Guidance Rewards with Trajectory-space Smoothing
• Deep Inductive Logic Programming meets Reinforcement Learning
• Linear Representation Meta-Reinforcement Learning for Instant Adaptation
• f-IRL: Inverse Reinforcement Learning via State Marginal Matching
• Offline Reinforcement Learning Hands-On
• Active Perception and Representation for Robotic Manipulation
• EdgeSlice: Slicing Wireless Edge Computing Network with Decentralized Deep Reinforcement Learning
• Iterative Amortized Policy Optimization
• Function Contrastive Learning of Transferable Meta-Representations
• Toddler-Guidance Learning: Impacts of Critical Period on Multimodal AI Agents
• Real-time Control of Electric Autonomous Mobility-on-Demand Systems via Graph Reinforcement Learning
• Mutual-Information Regularization in Markov Decision Processes and Actor-Critic Learning
• OER: Offline Experience Replay for Continual Offline Reinforcement Learning
• Towards Automatic Actor-Critic Solutions to Continuous Control
• Aggressive Q-Learning with Ensembles: Achieving Both High Sample Efficiency and High Asymptotic Performance
• Outcome-Driven Reinforcement Learning via Variational Inference
• Identifying Critical States by the Action-Based Variance of Expected Return
• Learning Value Functions in Deep Policy Gradients using Residual Variance
• On Effective Scheduling of Model-based Reinforcement Learning
• Certifiable Robustness to Adversarial State Uncertainty in Deep Reinforcement Learning
• Proximal Policy Gradient: PPO with Policy Gradient
• A Mapping of Assurance Techniques for Learning Enabled Autonomous Systems to the Systems Engineering Lifecycle
• Counterfactual Policy Evaluation for Decision-Making in Autonomous Driving
• Residual Policy Learning for Powertrain Control
• Towards Theoretical Understandings of Robust Markov Decision Processes: Sample Complexity and Asymptotics
• Experience Augmentation: Boosting and Accelerating Off-Policy Multi-Agent Reinforcement Learning
• PAC-Bayes Bounds for Bandit Problems: A Survey and Experimental Comparison
• Imitative Planning using Conditional Normalizing Flow
• Skill Discovery of Coordination in Multi-agent Reinforcement Learning
• GDI: Rethinking What Makes Reinforcement Learning Different From Supervised Learning
• Tutorial and Survey on Probabilistic Graphical Model and Variational Inference in Deep Reinforcement Learning
• Continuous Control for Searching and Planning with a Learned Model
• Dynamical Distance Learning for Semi-Supervised and Unsupervised Skill Discovery
• Globally Optimal Hierarchical Reinforcement Learning for Linearly-Solvable Markov Decision Processes
• MELD: Meta-Reinforcement Learning from Images via Latent State Models
• Conservative Data Sharing for Multi-Task Offline Reinforcement Learning
• DDPG++: Striving for Simplicity in Continuous-control Off-Policy Reinforcement Learning
• From Credit Assignment to Entropy Regularization: Two New Algorithms for Neural Sequence Prediction
• Control as Hybrid Inference
• Offline Decentralized Multi-Agent Reinforcement Learning
• Inverse Reinforcement Learning from a Gradient-based Learner
• Unsupervised Domain Adaptation with Dynamics-Aware Rewards in Reinforcement Learning
• WAD: A Deep Reinforcement Learning Agent for Urban Autonomous Driving
• Auto-Agent-Distiller: Towards Efficient Deep Reinforcement Learning Agents via Neural Architecture Search
• Economic Battery Storage Dispatch with Deep Reinforcement Learning from Rule-Based Demonstrations
• Inter-Level Cooperation in Hierarchical Reinforcement Learning
• Model-Invariant State Abstractions for Model-Based Reinforcement Learning
• Reinforcement Learning with Random Delays
• Reinforcement Learning in Non-Markov Market-Making
• Escaping from Zero Gradient: Revisiting Action-Constrained Reinforcement Learning via Frank-Wolfe Policy Optimization
• Distributed Soft Actor-Critic with Multivariate Reward Representation and Knowledge Distillation
• MobILE: Model-Based Imitation Learning From Observation Alone
• Behavior-Guided Actor-Critic: Improving Exploration via Learning Policy Behavior Representation for Deep Reinforcement Learning
• Uncertainty-Aware Model-Based Reinforcement Learning with Application to Autonomous Driving
• Momentum in Reinforcement Learning
• Merging Deterministic Policy Gradient Estimations with Varied Bias-Variance Tradeoff for Effective Deep Reinforcement Learning
• Estimating Optimal Infinite Horizon Dynamic Treatment Regimes via pT-Learning
• Divergence-Regularized Multi-Agent Actor-Critic
• Continuous Control with Action Quantization from Demonstrations
• Hindsight Foresight Relabeling for Meta-Reinforcement Learning
• Discriminator Soft Actor Critic without Extrinsic Rewards
• Towards Deeper Deep Reinforcement Learning with Spectral Normalization
• Pre-training of Deep RL Agents for Improved Learning under Domain Randomization
• PAC-Bayesian Soft Actor-Critic Learning
• Touch-based Curiosity for Sparse-Reward Tasks
• Understanding the Origin of Information-Seeking Exploration in Probabilistic Objectives for Control
• Peer Learning: Learning Complex Policies in Groups from Scratch via Action Recommendations
• Bias-reduced Multi-step Hindsight Experience Replay for Efficient Multi-goal Reinforcement Learning
• Building Minimal and Reusable Causal State Abstractions for Reinforcement Learning
• Provably Improved Context-Based Offline Meta-RL with Attention and Contrastive Learning
• Optimization of a Triangular Delaunay Mesh Generator using Reinforcement Learning
• Learning Pregrasp Manipulation of Objects from Ungraspable Poses
• Privacy-Constrained Policies via Mutual Information Regularized Policy Gradients
• Variational Transport: A Convergent Particle-BasedAlgorithm for Distributional Optimization
• Learning Diverse Policies with Soft Self-Generated Guidance
• SeRO: Self-Supervised Reinforcement Learning for Recovery from Out-of-Distribution Situations
• Regularized Inverse Reinforcement Learning
• Energy-based Surprise Minimization for Multi-Agent Value Factorization
• Linear Disentangled Representations and Unsupervised Action Estimation
• Boosting Offline Reinforcement Learning with Residual Generative Modeling
• Sample Efficient Reinforcement Learning via Model-Ensemble Exploration and Exploitation
• Embracing the Dark Knowledge: Domain Generalization Using Regularized Knowledge Distillation
• Safe Exploration by Solving Early Terminated MDP
• A First-Occupancy Representation for Reinforcement Learning
• Efficient Empowerment Estimation for Unsupervised Stabilization
• Learning to Sample with Local and Global Contexts in Experience Replay Buffer
• Optimizing Memory Placement using Evolutionary Graph Reinforcement Learning
• Landmark Policy Optimization for Object Navigation Task
• Zeroth-Order Supervised Policy Improvement
• Learning Temporally-Consistent Representations for Data-Efficient Reinforcement Learning
• Feedback Linearization of Car Dynamics for Racing via Reinforcement Learning
• Smooth Imitation Learning via Smooth Costs and Smooth Policies
• Exploration by Maximizing Rényi Entropy for Reward-Free RL Framework
• Finding the Near Optimal Policy via Adaptive Reduced Regularization in MDPs
• Novel Policy Seeking with Constrained Optimization
• Learning Trajectories for Visual-Inertial System Calibration via Model-based Heuristic Deep Reinforcement Learning
• Dynamic allocation of limited memory resources in reinforcement learning
• Diluted Near-Optimal Expert Demonstrations for Guiding Dialogue Stochastic Policy Optimisation
• Hierarchical Soft Actor-Critic: Adversarial Exploration via Mutual Information Optimization
• Fault-Aware Robust Control via Adversarial Reinforcement Learning
• Battery Model Calibration with Deep Reinforcement Learning
• Reinforcement Learning with Latent Flow
• Neural Lyapunov Model Predictive Control: Learning Safe Global Controllers from Sub-optimal Examples
• Value-driven Hindsight Modelling
• Mutual Information-based State-Control for Intrinsically Motivated Reinforcement Learning
• Policy Optimization Reinforcement Learning with Entropy Regularization
• Decision-Making under On-Ramp merge Scenarios by Distributional Soft Actor-Critic Algorithm
• Self-Imitation Learning by Planning
• Sample Efficient Ensemble Learning with Catalyst.RL
• Forgetful Experience Replay in Hierarchical Reinforcement Learning from Demonstrations
• Maximum Entropy Model Rollouts: Fast Model Based Policy Optimization without Compounding Errors
• Simulating multi-exit evacuation using deep reinforcement learning
• Market-making with reinforcement-learning (SAC)
• Approximate Inference in Discrete Distributions with Monte Carlo Tree Search and Value Functions
• Deep Model-Based Reinforcement Learning via Estimated Uncertainty and Conservative Policy Optimization
• A Free Lunch from the Noise: Provable and Practical Exploration for Representation Learning
• Learning State Representations via Retracing in Reinforcement Learning
• A Generalised Inverse Reinforcement Learning Framework
• C-Learning: Horizon-Aware Cumulative Accessibility Estimation
• Discrete Action On-Policy Learning with Action-Value Critic
• Which Channel to Ask My Question? Personalized Customer Service Request Stream Routing using Deep Reinforcement Learning
• Logistic Q-Learning
• GloCAL: Glocalized Curriculum-Aided Learning of Multiple Tasks with Application to Robotic Grasping
• KSG: Knowledge and Skill Graph
• Self-Paced Deep Reinforcement Learning
• Learning to Plan Optimistically: Uncertainty-Guided Deep Exploration via Latent Model Ensembles
• PFPN: Continuous Control of Physically Simulated Characters using Particle Filtering Policy Network
• A survey of benchmarking frameworks for reinforcement learning
• Reinforcement Learning in the Wild with Maximum Likelihood-based Model Transfer
• Ground States of Quantum Many Body Lattice Models via Reinforcement Learning
• RTDK-BO: High Dimensional Bayesian Optimization with Reinforced Transformer Deep kernels
• Dream and Search to Control: Latent Space Planning for Continuous Control
• On the Sensory Commutativity of Action Sequences for Embodied Agents
• Distill Knowledge in Multi-task Reinforcement Learning with Optimal-Transport Regularization
• Regularized Policies are Reward Robust
• Representation of Reinforcement Learning Policies in Reproducing Kernel Hilbert Spaces
• Learning task-agnostic representation via toddler-inspired learning
• Measuring Progress in Deep Reinforcement Learning Sample Efficiency
• Tackling Diverse Minorities in Imbalanced Classification
• Structured Diversification Emergence via Reinforced Organization Control and Hierarchical Consensus Learning
• Function Approximation for Reinforcement Learning Controller for Energy from Spread Waves
• Planning under Uncertainty to Goal Distributions
• Symmetric Q-learning: Reducing Skewness of Bellman Error in Online Reinforcement Learning
• Distilling Motion Planner Augmented Policies into Visual Control Policies for Robot Manipulation
• Action Redundancy in Reinforcement Learning
• Path Following and Stabilisation of a Bicycle Model using a Reinforcement Learning Approach
• Curriculum Offline Imitation Learning
• MetricOpt: Learning to Optimize Black-Box Evaluation Metrics
• On the Expressivity of Neural Networks for Deep Reinforcement Learning
• Learning Resilient Behaviors for Navigation Under Uncertainty
• Improving Reinforcement Learning Efficiency with Auxiliary Tasks in Non-Visual Environments: A Comparison
• MDP Playground: An Analysis and Debug Testbed for Reinforcement Learning
• Tuning Synaptic Connections instead of Weights by Genetic Algorithm in Spiking Policy Network
• ISL: A novel approach for deep exploration
• Semi-On-Policy Training for Sample Efficient Multi-Agent Policy Gradients
• Advanced Policies: A First-Principles Path from Policy Gradient to Q-Learning
• Reinforcement learning with world model
• On the Convergence Rate of Off-Policy Policy Optimization Methods with Density-Ratio Correction
• Analysis of a Target-Based Actor-Critic Algorithm with Linear Function Approximation
• A Closer Look at Advantage-Filtered Behavioral Cloning in High-Noise Datasets
• Learning MDPs from Features: Predict-Then-Optimize for Sequential Decision Problems by Reinforcement Learning
• IQ-Learn: Inverse soft-Q Learning for Imitation
• Intrinsically motivated option learning: a comparative study of recent methods
• Learning Altruistic Behaviours in Reinforcement Learning without External Rewards
• Off-policy Reinforcement Learning with Optimistic Exploration and Distribution Correction
• Regularly Updated Deterministic Policy Gradient Algorithm
• Supervised Off-Policy Ranking
• Reinforcement Learning as Iterative and Amortised Inference
• Learning Policies through Quantile Regression
• Reinforcement Learning for Vision-based Object Manipulation with Non-parametric Policy and Action Primitives
• Koopman Q-learning: Offline Reinforcement Learning via Symmetries of Dynamics
• Battlesnake Challenge: A Multi-agent Reinforcement Learning Playground with Human-in-the-loop
• Investigation of Factorized Optical Flows as Mid-Level Representations
• Implicitly Regularized RL with Implicit Q-Values
• Potential-Based Advice for Stochastic Policy Learning
• Improved Robustness and Safety for Pre-Adaptation of Meta Reinforcement Learning with Prior Regularization
• Self-supervised Reinforcement Learning with Independently Controllable Subgoals
• Learning Representations for Pixel-based Control: What Matters and Why?
• Soft Actor-Critic With Integer Actions
• LDC-VAE: A Latent Distribution Consistency Approach to Variational AutoEncoders
• Lyapunov-Based Reinforcement Learning for Decentralized Multi-Agent Control
• ADER:Adapting between Exploration and Robustness for Actor-Critic Methods
• Continuous Control With Ensemble Deep Deterministic Policy Gradients
• Adaptive Symmetric Reward Noising for Reinforcement Learning
• Importance Sampling based Exploration in Q Learning
• Convex Regularization in Monte-Carlo Tree Search
• Reinforcement Learning for High-dimensional Continuous Control in Biomechanics: An Intro to ArtiSynth-RL
• Recomposing the Reinforcement Learning Building Blocks with Hypernetworks
• Parameter-Based Value Functions
• On the Convergence of Approximate and Regularized Policy Iteration Schemes
• Concave Utility Reinforcement Learning with Zero-Constraint Violations
• A Self-adaptive LSAC-PID Approach based on Lyapunov Reward Shaping for Mobile Robots
• Offline RL With Resource Constrained Online Deployment
• An Entropy Regularization Free Mechanism for Policy-based Reinforcement Learning
• Deep Reinforcement Learning for Field Development Optimization
• Parameterized MDPs and Reinforcement Learning Problems -- A Maximum Entropy Principle Based Framework
• Co-Adaptation of Algorithmic and Implementational Innovations in Inference-based Deep Reinforcement Learning
• MBDP: A Model-based Approach to Achieve both Robustness and Sample Efficiency via Double Dropout Planning
• Countering the Effects of Lead Bias in News Summarization via Multi-Stage Training and Auxiliary Losses
• Prioritized Soft Q-Decomposition for Lexicographic Reinforcement Learning
• Large Batch Experience Replay
• Trying AGAIN instead of Trying Longer: Prior Learning for Automatic Curriculum Learning
• Homotopy Based Reinforcement Learning with Maximum Entropy for Autonomous Air Combat
• Model Predictive Actor-Critic: Accelerating Robot Skill Acquisition with Deep Reinforcement Learning
• Learning Efficient and Effective Exploration Policies with Counterfactual Meta Policy
• End-to-end grasping policies for human-in-the-loop robots via deep reinforcement learning
• Generative Actor-Critic: An Off-policy Algorithm Using the Push-forward Model
• Reinforcement Learning With Sparse-Executing Actions via Sparsity Regularization
• Learning What To Do by Simulating the Past
• Efficient Reinforcement Learning Development with RLzoo
• Direct Random Search for Fine Tuning of Deep Reinforcement Learning Policies
• Utilizing Skipped Frames in Action Repeats via Pseudo-Actions
• RAIL: A modular framework for Reinforcement-learning-based Adversarial Imitation Learning
• Scalable, Decentralized Multi-Agent Reinforcement Learning Methods Inspired by Stigmergy and Ant Colonies
• Neural Embedding for Physical Manipulations
• Policy-Based Reinforcement Learning for Assortative Matching in Human Behavior Modeling
• Joint Perception and Control as Inference with an Object-based Implementation
• Leveraging Sequentiality in Reinforcement Learning from a Single Demonstration
• From proprioception to long-horizon planning in novel environments: A hierarchical RL model
• Is High Variance Unavoidable in RL? A Case Study in Continuous Control
• A Contraction Approach to Model-based Reinforcement Learning
• Transfer Learning by Modeling a Distribution over Policies
• Evaluating task-agnostic exploration for fixed-batch learning of arbitrary future tasks
• Learning to Reach, Swim, Walk and Fly in One Trial: Data-Driven Control with Scarce Data and Side Information
• Optimizing Quantum Variational Circuits with Deep Reinforcement Learning
• Towards a Sample Efficient Reinforcement Learning Pipeline for Vision Based Robotics
• On-Policy Model Errors in Reinforcement Learning
• A Geometric Perspective on Self-Supervised Policy Adaptation
• Weighted Entropy Modification for Soft Actor-Critic
• Attention-based Fault-tolerant Approach for Multi-agent Reinforcement Learning Systems
• Model Embedding Model-Based Reinforcement Learning
• Planning with Exploration: Addressing Dynamics Bottleneck in Model-based Reinforcement Learning
• Towards Learning Controllable Representations of Physical Systems
• Batch Inverse-Variance Weighting: Deep Heteroscedastic Regression
• Learning of feature points without additional supervision improves reinforcement learning from images
• Contrastive Active Inference
• Entropy based Independent Learning in Anonymous Multi-Agent Settings
• Placement Optimization with Deep Reinforcement Learning
• Rethinking Uncertainty in Deep Learning: Whether and How it Improves Robustness
• Generative Temporal Difference Learning for Infinite-Horizon Prediction
• Mixture-of-Variational-Experts for Continual Learning
• Adaptation of Quadruped Robot Locomotion with Meta-Learning
• Real-time Flexibility Feedback for Closed-loop Aggregator and System Operator Coordination
• FiDi-RL: Incorporating Deep Reinforcement Learning with Finite-Difference Policy Search for Efficient Learning of Continuous Control
• Training Transition Policies via Distribution Matching for Complex Tasks
• On the Guaranteed Almost Equivalence between Imitation Learning from Observation and Demonstration
• Solving the Real Robot Challenge using Deep Reinforcement Learning
• Cautious Policy Programming: Exploiting KL Regularization in Monotonic Policy Improvement for Reinforcement Learning
• Proximal Policy Optimization with Mixed Distributed Training
• On The Transferability of Deep-Q Networks
• Imitation with Neural Density Models
• Implicit recurrent networks: A novel approach to stationary input processing with recurrent neural networks in deep learning
• OffCon${}^3$: What is state of the art anyway?
• Crowdfunding Dynamics Tracking: A Reinforcement Learning Approach
• Direct then Diffuse: Incremental Unsupervised Skill Discovery for State Covering and Goal Reaching
• CoachNet: An Adversarial Sampling Approach for Reinforcement Learning
• Canonical Cortical Circuits and the Duality of Bayesian Inference and Optimal Control
• Deep Reinforcement Learning with Adjustments
• Policy Gradient RL Algorithms as Directed Acyclic Graphs
• Showing Your Offline Reinforcement Learning Work: Online Evaluation Budget Matters
• Continuous Multi-objective Zero-touch Network Slicing via Twin Delayed DDPG and OpenAI Gym
• Developing cooperative policies for multi-stage tasks
• RLCache: Automated Cache Management Using Reinforcement Learning
• Global optimality of softmax policy gradient with single hidden layer neural networks in the mean-field regime
• Reinforcement Learning for Robust Missile Autopilot Design
• Offline Inverse Reinforcement Learning
• Uncertainty-aware Low-Rank Q-Matrix Estimation for Deep Reinforcement Learning
• Sharp Analysis of Smoothed Bellman Error Embedding
• Expert Human-Level Driving in Gran Turismo Sport Using Deep Reinforcement Learning with Image-based Representation
• Ranking Policy Gradient
• Can Q-learning solve Multi Armed Bantids?
• Quickest change detection with unknown parameters: Constant complexity and near optimality
• Mix and Mask Actor-Critic Methods
• Consolidation via Policy Information Regularization in Deep RL for Multi-Agent Games
• How to Train your Quadrotor: A Framework for Consistently Smooth and Responsive Flight Control via Reinforcement Learning
• Decoupled Exploration and Exploitation Policies for Sample-Efficient Reinforcement Learning
• Zero-shot generalization using cascaded system-representations
• Medical Vision Language Models as Policies for Robotic Surgery
• Global Optimality and Finite Sample Analysis of Softmax Off-Policy Actor Critic under State Distribution Mismatch
• Variational Policy Search using Sparse Gaussian Process Priors for Learning Multimodal Optimal Actions
• Learning to See via Epiretinal Implant Stimulation in silico with Model-Based Deep Reinforcement Learning
• Reinforcement Explanation Learning
• Enhancing Reinforcement Learning Through Guided Search
• OpEvo: An Evolutionary Method for Tensor Operator Optimization
• Information Directed Reward Learning for Reinforcement Learning
• Concurrent Training Improves the Performance of Behavioral Cloning from Observation
• State Representation Learning from Demonstration
• Maximum Entropy Reinforcement Learning with Mixture Policies
• Towards an Understanding of Default Policies in Multitask Policy Optimization
• Hierarchical and Partially Observable Goal-driven Policy Learning with Goals Relational Graph
• Generalizing Decision Making for Automated Driving with an Invariant Environment Representation using Deep Reinforcement Learning
• Planning and Learning Using Adaptive Entropy Tree Search
• Reinforcement Learning for Orientation Estimation Using Inertial Sensors with Performance Guarantee
• Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation Gap
• SafePILCO: a software tool for safe and data-efficient policy synthesis