Soft Actor-Critic Algorithms and Applications

Abstract

Model-free deep reinforcement learning (RL) algorithms have been successfully applied to a range of challenging sequential decision making and control tasks. However, these methods typically suffer from two major challenges: high sample complexity and brittleness to hyperparameters. Both of these challenges limit the applicability of such methods to real-world domains. In this paper, we describe Soft Actor-Critic (SAC), our recently introduced off-policy actor-critic algorithm based on the maximum entropy RL framework. In this framework, the actor aims to simultaneously maximize expected return and entropy. That is, to succeed at the task while acting as randomly as possible. We extend SAC to incorporate a number of modifications that accelerate training and improve stability with respect to the hyperparameters, including a constrained formulation that automatically tunes the temperature hyperparameter. We systematically evaluate SAC on a range of benchmark tasks, as well as real-world challenging tasks such as locomotion for a quadrupedal robot and robotic manipulation with a dexterous hand. With these improvements, SAC achieves state-of-the-art performance, outperforming prior on-policy and off-policy methods in sample-efficiency and asymptotic performance. Furthermore, we demonstrate that, in contrast to other off-policy algorithms, our approach is very stable, achieving similar performance across different random seeds. These results suggest that SAC is a promising candidate for learning in real-world robotics tasks.

arXiv admin note: substantial text overlap with arXiv:1801.01290

References in corpus (15)

• Continuous control with deep reinforcement learning
• Playing Atari with Deep Reinforcement Learning
• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Trust Region Policy Optimization
• Addressing Function Approximation Error in Actor-Critic Methods
• Reinforcement Learning with Deep Energy-Based Policies
• Learning Continuous Control Policies by Stochastic Value Gradients
• Path integrals and symmetry breaking for optimal control theory
• Maximum a Posteriori Policy Optimisation
• Bridging the Gap Between Value and Policy Based Reinforcement Learning
• Equivalence Between Policy Gradients and Soft Q-Learning
• 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
• Dexterous Manipulation with Deep Reinforcement Learning: Efficient, General, and Low-Cost

Cited by in corpus (327)

• Social physics
• Solving Rubik's Cube with a Robot Hand
• How to Train Your Robot with Deep Reinforcement Learning; Lessons We've Learned
• Distributional Soft Actor-Critic: Off-Policy Reinforcement Learning for Addressing Value Estimation Errors
• A Survey of Deep RL and IL for Autonomous Driving Policy Learning
• CURL: Contrastive Unsupervised Representations for Reinforcement Learning
• Soft Actor-Critic for Discrete Action Settings
• robosuite: A Modular Simulation Framework and Benchmark for Robot Learning
• RLOC: Terrain-Aware Legged Locomotion using Reinforcement Learning and Optimal Control
• Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
• Identifying optimal cycles in quantum thermal machines with reinforcement-learning
• What Matters In On-Policy Reinforcement Learning? A Large-Scale Empirical Study
• Learning to Walk in the Real World with Minimal Human Effort
• Automated Reinforcement Learning (AutoRL): A Survey and Open Problems
• Goal-Guided Transformer-Enabled Reinforcement Learning for Efficient Autonomous Navigation
• Learning Predictive Representations for Deformable Objects Using Contrastive Estimation
• Approximate Convex Decomposition for 3D Meshes with Collision-Aware Concavity and Tree Search
• ChainerRL: A Deep Reinforcement Learning Library
• Dynamics-Aware Unsupervised Discovery of Skills
• Adaptive dynamic programming for nonaffine nonlinear optimal control problem with state constraints
• Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learning
• Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model
• Self-Supervised Policy Adaptation during Deployment
• MERLIN: Multi-agent offline and transfer learning for occupant-centric energy flexible operation of grid-interactive communities using smart meter data and CityLearn
• Reinforcement learning for automatic quadrilateral mesh generation: a soft actor-critic approach
• Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics
• Real-world challenges for multi-agent reinforcement learning in grid-interactive buildings
• Residual Policy Learning
• rlpyt: A Research Code Base for Deep Reinforcement Learning in PyTorch
• Reinforcement Learning for Generative AI: State of the Art, Opportunities and Open Research Challenges
• RecSim: A Configurable Simulation Platform for Recommender Systems
• An empirical investigation of the challenges of real-world reinforcement learning
• Collaborative Optimization of Multi-microgrids System with Shared Energy Storage Based on Multi-agent Stochastic Game and Reinforcement Learning
• Keep Doing What Worked: Behavioral Modelling Priors for Offline Reinforcement Learning
• Off-Policy Policy Gradient with State Distribution Correction
• Effective Diversity in Population Based Reinforcement Learning
• iGibson 1.0: a Simulation Environment for Interactive Tasks in Large Realistic Scenes
• d3rlpy: An Offline Deep Reinforcement Learning Library
• Robust Reinforcement Learning for Continuous Control with Model Misspecification
• Reinforcement Learning with Prototypical Representations
• Model-Augmented Actor-Critic: Backpropagating through Paths
• Boosting Soft Actor-Critic: Emphasizing Recent Experience without Forgetting the Past
• Diversity Policy Gradient for Sample Efficient Quality-Diversity Optimization
• Online Multi-agent Reinforcement Learning for Decentralized Inverter-based Volt-VAR Control
• Differentiated Federated Reinforcement Learning Based Traffic Offloading on Space-Air-Ground Integrated Networks
• Discrete and Continuous Action Representation for Practical RL in Video Games
• ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulation
• Sim-to-Real Surgical Robot Learning and Autonomous Planning for Internal Tissue Points Manipulation using Reinforcement Learning
• Learning Purely Tactile In-Hand Manipulation with a Torque-Controlled Hand
• Primal Wasserstein Imitation Learning
• COG: Connecting New Skills to Past Experience with Offline Reinforcement Learning
• Reinforcement learning of rare diffusive dynamics
• A Deep Reinforcement Learning-Based Resource Scheduler for Massive MIMO Networks
• Automatic Curriculum Learning through Value Disagreement
• A Self-Tuning Actor-Critic Algorithm
• BAIL: Best-Action Imitation Learning for Batch Deep Reinforcement Learning
• Learning from Suboptimal Demonstration via Self-Supervised Reward Regression
• Model-free Deep Reinforcement Learning for Urban Autonomous Driving
• Quantum imaginary time evolution steered by reinforcement learning
• Rewriting History with Inverse RL: Hindsight Inference for Policy Improvement
• Imitation Learning via Off-Policy Distribution Matching
• Deep Reinforcement Learning for Closed-Loop Blood Glucose Control
• Trajectory Planning with Deep Reinforcement Learning in High-Level Action Spaces
• Parrot: Data-Driven Behavioral Priors for Reinforcement Learning
• ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrations
• Continuous-Discrete Reinforcement Learning for Hybrid Control in Robotics
• MCI-GRU: Stock Prediction Model Based on Multi-Head Cross-Attention and Improved GRU
• The Ingredients of Real-World Robotic Reinforcement Learning
• Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
• Learning to be Safe: Deep RL with a Safety Critic
• EfficientGrasp: A Unified Data-Efficient Learning to Grasp Method for Multi-fingered Robot Hands
• Offline Reinforcement Learning with Fisher Divergence Critic Regularization
• Interpretable End-to-end Urban Autonomous Driving with Latent Deep Reinforcement Learning
• Knowledge Transfer for Cross-Domain Reinforcement Learning: A Systematic Review
• Predictive Information Accelerates Learning in RL
• Forward and inverse reinforcement learning sharing network weights and hyperparameters
• Evolving Rewards to Automate Reinforcement Learning
• Soft Actor-Critic Deep Reinforcement Learning for Fault Tolerant Flight Control
• Learning Latent Plans from Play
• Representation Matters: Offline Pretraining for Sequential Decision Making
• DoorGym: A Scalable Door Opening Environment And Baseline Agent
• A Survey on Reinforcement Learning for Combinatorial Optimization
• CACTO: Continuous Actor-Critic with Trajectory Optimization -- Towards global optimality
• Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulation
• Learning to Seek: Autonomous Source Seeking with Deep Reinforcement Learning Onboard a Nano Drone Microcontroller
• Continual World: A Robotic Benchmark For Continual Reinforcement Learning
• Variational Recurrent Models for Solving Partially Observable Control Tasks
• Hit and Lead Discovery with Explorative RL and Fragment-based Molecule Generation
• Recovery RL: Safe Reinforcement Learning with Learned Recovery Zones
• Reinforced Neighborhood Selection Guided Multi-Relational Graph Neural Networks
• Learning to Grasp on the Moon from 3D Octree Observations with Deep Reinforcement Learning
• Dropout Q-Functions for Doubly Efficient Reinforcement Learning
• Online Meta-Critic Learning for Off-Policy Actor-Critic Methods
• Learning to Manipulate Deformable Objects without Demonstrations
• Meta-SAC: Auto-tune the Entropy Temperature of Soft Actor-Critic via Metagradient
• Generalized Hindsight for Reinforcement Learning
• Imitation Learning for High Precision Peg-in-Hole Tasks
• Sparse Graphical Memory for Robust Planning
• What Matters for Adversarial Imitation Learning?
• Distributionally Robust Reinforcement Learning
• Dynamics-aware Embeddings
• UnifiedGesture: A Unified Gesture Synthesis Model for Multiple Skeletons
• A Safe Deep Reinforcement Learning Approach for Energy Efficient Federated Learning in Wireless Communication Networks
• A Unified Bellman Optimality Principle Combining Reward Maximization and Empowerment
• On the model-based stochastic value gradient for continuous reinforcement learning
• Deep Reinforcement Learning for Time-Critical Wilderness Search And Rescue Using Drones
• Towards Real-World Applications of Personalized Anesthesia Using Policy Constraint Q Learning for Propofol Infusion Control
• STAP: Sequencing Task-Agnostic Policies
• Learning Visual Robotic Control Efficiently with Contrastive Pre-training and Data Augmentation
• SECANT: Self-Expert Cloning for Zero-Shot Generalization of Visual Policies
• Off-Policy Reinforcement Learning with Delayed Rewards
• Multi-agent Reinforcement Learning Accelerated MCMC on Multiscale Inversion Problem
• Hallucinative Topological Memory for Zero-Shot Visual Planning
• RLAD: Reinforcement Learning from Pixels for Autonomous Driving in Urban Environments
• Asynchronous Methods for Model-Based Reinforcement Learning
• Deep reinforcement learning under signal temporal logic constraints using Lagrangian relaxation
• Orchestrating the Development Lifecycle of Machine Learning-Based IoT Applications: A Taxonomy and Survey
• Recurrent Off-policy Baselines for Memory-based Continuous Control
• A Versatile Door Opening System with Mobile Manipulator through Adaptive Position-Force Control and Reinforcement Learning
• Bag of Views: An Appearance-based Approach to Next-Best-View Planning for 3D Reconstruction
• Striving for Simplicity and Performance in Off-Policy DRL: Output Normalization and Non-Uniform Sampling
• F2A2: Flexible Fully-decentralized Approximate Actor-critic for Cooperative Multi-agent Reinforcement Learning
• Real-Time Model Calibration with Deep Reinforcement Learning
• CCLF: A Contrastive-Curiosity-Driven Learning Framework for Sample-Efficient Reinforcement Learning
• Muesli: Combining Improvements in Policy Optimization
• Learning Obstacle Representations for Neural Motion Planning
• Learning to simulate and design for structural engineering
• A Tutorial on Sparse Gaussian Processes and Variational Inference
• Efficient Deep Reinforcement Learning with Imitative Expert Priors for Autonomous Driving
• PowerGym: A Reinforcement Learning Environment for Volt-Var Control in Power Distribution Systems
• ROBEL: Robotics Benchmarks for Learning with Low-Cost Robots
• An Equivalence between Loss Functions and Non-Uniform Sampling in Experience Replay
• Skip Training for Multi-Agent Reinforcement Learning Controller for Industrial Wave Energy Converters
• XIRL: Cross-embodiment Inverse Reinforcement Learning
• MAT: Multi-Fingered Adaptive Tactile Grasping via Deep Reinforcement Learning
• Model-Reference Reinforcement Learning Control of Autonomous Surface Vehicles with Uncertainties
• Emergent Real-World Robotic Skills via Unsupervised Off-Policy Reinforcement Learning
• Off-policy Maximum Entropy Reinforcement Learning : Soft Actor-Critic with Advantage Weighted Mixture Policy(SAC-AWMP)
• Actor-Critic Reinforcement Learning for Control with Stability Guarantee
• Demonstration-Guided Reinforcement Learning with Learned Skills
• Augmented World Models Facilitate Zero-Shot Dynamics Generalization From a Single Offline Environment
• Continual Reinforcement Learning with Multi-Timescale Replay
• Noise-Robust End-to-End Quantum Control using Deep Autoregressive Policy Networks
• Diversity Actor-Critic: Sample-Aware Entropy Regularization for Sample-Efficient Exploration
• Smooth Exploration for Robotic Reinforcement Learning
• Thinking While Moving: Deep Reinforcement Learning with Concurrent Control
• A Grasp Pose is All You Need: Learning Multi-fingered Grasping with Deep Reinforcement Learning from Vision and Touch
• Offline Meta-Reinforcement Learning with Online Self-Supervision
• To Follow or not to Follow: Selective Imitation Learning from Observations
• Optimizing for the Future in Non-Stationary MDPs
• On Entropy Regularized Path Integral Control for Trajectory Optimization
• Guided Policy Search Model-based Reinforcement Learning for Urban Autonomous Driving
• PlayVirtual: Augmenting Cycle-Consistent Virtual Trajectories for Reinforcement Learning
• Measuring Visual Generalization in Continuous Control from Pixels
• Learning-based Initialization of Trajectory Optimization for Path-following Problems of Redundant Manipulators
• Reinforcement Learning Generalization with Surprise Minimization
• Ready Policy One: World Building Through Active Learning
• Interactive Imitation Learning for Dexterous Robotic Manipulation: Challenges and Perspectives -- A Survey
• Adversarial Soft Advantage Fitting: Imitation Learning without Policy Optimization
• Guidance Design for Escape Flight Vehicle Using Evolution Strategy Enhanced Deep Reinforcement Learning
• A Decentralized Policy Gradient Approach to Multi-task Reinforcement Learning
• SOAC: The Soft Option Actor-Critic Architecture
• Exploratory State Representation Learning
• MARLIN: Soft Actor-Critic based Reinforcement Learning for Congestion Control in Real Networks
• Deep Reinforcement Learning for Autonomous Ground Vehicle Exploration Without A-Priori Maps
• A Framework for Studying Reinforcement Learning and Sim-to-Real in Robot Soccer
• Deep Reinforcement Learning-based UAV Navigation and Control: A Soft Actor-Critic with Hindsight Experience Replay Approach
• SLM Lab: A Comprehensive Benchmark and Modular Software Framework for Reproducible Deep Reinforcement Learning
• Dynamic Queue-Jump Lane for Emergency Vehicles under Partially Connected Settings: A Multi-Agent Deep Reinforcement Learning Approach
• Two-stage Deep Reinforcement Learning for Inverter-based Volt-VAR Control in Active Distribution Networks
• Certificated Actor-Critic: Hierarchical Reinforcement Learning with Control Barrier Functions for Safe Navigation
• Sim-to-Real Transfer of Deep Reinforcement Learning Agents for Online Coverage Path Planning
• Colored Noise in PPO: Improved Exploration and Performance through Correlated Action Sampling
• Model-Reference Reinforcement Learning for Collision-Free Tracking Control of Autonomous Surface Vehicles
• Reinforcement Learning by Guided Safe Exploration
• Exploration-efficient Deep Reinforcement Learning with Demonstration Guidance for Robot Control
• Augmenting learning using symmetry in a biologically-inspired domain
• TAAC: Temporally Abstract Actor-Critic for Continuous Control
• Imagined Value Gradients: Model-Based Policy Optimization with Transferable Latent Dynamics Models
• An Adaptable Approach to Learn Realistic Legged Locomotion without Examples
• Learning Barrier Certificates: Towards Safe Reinforcement Learning with Zero Training-time Violations
• Model-free and Bayesian Ensembling Model-based Deep Reinforcement Learning for Particle Accelerator Control Demonstrated on the FERMI FEL
• Reinforcement Learning with Videos: Combining Offline Observations with Interaction
• Conformal Symplectic Optimization for Stable Reinforcement Learning
• Toolpath design for additive manufacturing using deep reinforcement learning
• Deep RL With Information Constrained Policies: Generalization in Continuous Control
• A Max-Min Entropy Framework for Reinforcement Learning
• Counterfactual experience augmented off-policy reinforcement learning
• Hamilton-Jacobi-Bellman Equations for Maximum Entropy Optimal Control
• Bi-level Off-policy Reinforcement Learning for Volt/VAR Control Involving Continuous and Discrete Devices
• Learning Object-conditioned Exploration using Distributed Soft Actor Critic
• GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning
• Implicit Distributional Reinforcement Learning
• Encoding Distributional Soft Actor-Critic for Autonomous Driving in Multi-lane Scenarios
• Caching Transient Content for IoT Sensing: Multi-Agent Soft Actor-Critic
• Using Soft Actor-Critic for Low-Level UAV Control
• Learning Generalizable Locomotion Skills with Hierarchical Reinforcement Learning
• Benchmarking Structured Policies and Policy Optimization for Real-World Dexterous Object Manipulation
• Iterative Amortized Policy Optimization
• AW-Opt: Learning Robotic Skills with Imitation and Reinforcement at Scale
• Learning Long-Term Reward Redistribution via Randomized Return Decomposition
• Active Perception and Representation for Robotic Manipulation
• RLlib Flow: Distributed Reinforcement Learning is a Dataflow Problem
• Off-Policy Adversarial Inverse Reinforcement Learning
• Unsupervised Visual Attention and Invariance for Reinforcement Learning
• Solving Challenging Dexterous Manipulation Tasks With Trajectory Optimisation and Reinforcement Learning
• Reinforcement Learning with Structured Hierarchical Grammar Representations of Actions
• Domain-Robust Visual Imitation Learning with Mutual Information Constraints
• Visual-Policy Learning through Multi-Camera View to Single-Camera View Knowledge Distillation for Robot Manipulation Tasks
• Deep Reinforcement Learning for Process Synthesis
• Economic Battery Storage Dispatch with Deep Reinforcement Learning from Rule-Based Demonstrations
• Variational Empowerment as Representation Learning for Goal-Based Reinforcement Learning
• Reinforcement Learning Control of Constrained Dynamic Systems with Uniformly Ultimate Boundedness Stability Guarantee
• Goal-Directed Planning by Reinforcement Learning and Active Inference
• A Model-Based Approach for Improving Reinforcement Learning Efficiency Leveraging Expert Observations
• Continuous Transition: Improving Sample Efficiency for Continuous Control Problems via MixUp
• Uncertainty-Aware Model-Based Reinforcement Learning with Application to Autonomous Driving
• Revisiting Prioritized Experience Replay: A Value Perspective
• T3S: Improving Multi-Task Reinforcement Learning with Task-Specific Feature Selector and Scheduler
• Hierarchically Decoupled Imitation for Morphological Transfer
• Outcome-Driven Reinforcement Learning via Variational Inference
• Is Bang-Bang Control All You Need? Solving Continuous Control with Bernoulli Policies
• Dynamical Distance Learning for Semi-Supervised and Unsupervised Skill Discovery
• Exploring the robustness of TractOracle methods in RL-based tractography
• Abstract Demonstrations and Adaptive Exploration for Efficient and Stable Multi-step Sparse Reward Reinforcement Learning
• Aggressive Q-Learning with Ensembles: Achieving Both High Sample Efficiency and High Asymptotic Performance
• Learning Latent State Spaces for Planning through Reward Prediction
• OER: Offline Experience Replay for Continual Offline Reinforcement Learning
• Mutual-Information Regularization in Markov Decision Processes and Actor-Critic Learning
• Cross-Domain Imitation Learning via Optimal Transport
• Distributed Soft Actor-Critic with Multivariate Reward Representation and Knowledge Distillation
• On the Privacy Risks of Deploying Recurrent Neural Networks in Machine Learning Models
• Towards Automatic Actor-Critic Solutions to Continuous Control
• Blending MPC & Value Function Approximation for Efficient Reinforcement Learning
• Reinforcement Learning with Random Delays
• Multi-Modal Mutual Information (MuMMI) Training for Robust Self-Supervised Deep Reinforcement Learning
• A Tensor Network Approach to Finite Markov Decision Processes
• Discriminator Soft Actor Critic without Extrinsic Rewards
• Maximum Entropy Diverse Exploration: Disentangling Maximum Entropy Reinforcement Learning
• Market-making with reinforcement-learning (SAC)
• A Relation Analysis of Markov Decision Process Frameworks
• OCEAN: Online Task Inference for Compositional Tasks with Context Adaptation
• Maximum Entropy Model Rollouts: Fast Model Based Policy Optimization without Compounding Errors
• Self-Imitation Learning by Planning
• Improving Context-Based Meta-Reinforcement Learning with Self-Supervised Trajectory Contrastive Learning
• Decision-Making under On-Ramp merge Scenarios by Distributional Soft Actor-Critic Algorithm
• OPAC: Opportunistic Actor-Critic
• Battery Model Calibration with Deep Reinforcement Learning
• Bridging Scene Understanding and Task Execution with Flexible Simulation Environments
• Temporal-Difference Value Estimation via Uncertainty-Guided Soft Updates
• RoMA: Robust Model Adaptation for Offline Model-based Optimization
• Learning Temporally-Consistent Representations for Data-Efficient Reinforcement Learning
• Improved Soft Actor-Critic: Mixing Prioritized Off-Policy Samples with On-Policy Experience
• Learning to Navigate Sidewalks in Outdoor Environments
• A Simple Unified Uncertainty-Guided Framework for Offline-to-Online Reinforcement Learning
• Learn Zero-Constraint-Violation Policy in Model-Free Constrained Reinforcement Learning
• Double Deep Q-learning Based Real-Time Optimization Strategy for Microgrids
• Understanding the Origin of Information-Seeking Exploration in Probabilistic Objectives for Control
• Policy Information Capacity: Information-Theoretic Measure for Task Complexity in Deep Reinforcement Learning
• Unlocking Pixels for Reinforcement Learning via Implicit Attention
• Learning to Sample with Local and Global Contexts in Experience Replay Buffer
• Policy Optimization Reinforcement Learning with Entropy Regularization
• Learning Altruistic Behaviours in Reinforcement Learning without External Rewards
• Investigation of Factorized Optical Flows as Mid-Level Representations
• Learning Off-Policy with Online Planning
• Next-Best-View Estimation based on Deep Reinforcement Learning for Active Object Classification
• Hierarchical Skills for Efficient Exploration
• Distributed Multi-Agent Deep Reinforcement Learning Framework for Whole-building HVAC Control
• Towards Robust Bisimulation Metric Learning
• What Robot do I Need? Fast Co-Adaptation of Morphology and Control using Graph Neural Networks
• On Training Effective Reinforcement Learning Agents for Real-time Power Grid Operation and Control
• Implicitly Regularized RL with Implicit Q-Values
• On the Convergence of Approximate and Regularized Policy Iteration Schemes
• Periodic Intra-Ensemble Knowledge Distillation for Reinforcement Learning
• Reducing Conservativeness Oriented Offline Reinforcement Learning
• The AI Arena: A Framework for Distributed Multi-Agent Reinforcement Learning
• Using Logical Specifications of Objectives in Multi-Objective Reinforcement Learning
• Near Optimal Policy Optimization via REPS
• Learning Pessimism for Robust and Efficient Off-Policy Reinforcement Learning
• Model-based Multi-agent Policy Optimization with Adaptive Opponent-wise Rollouts
• Towards neural reinforcement learning for large deviations in nonequilibrium systems with memory
• Bellman: A Toolbox for Model-Based Reinforcement Learning in TensorFlow
• Action Candidate Based Clipped Double Q-learning for Discrete and Continuous Action Tasks
• Reinforcement Learning as Iterative and Amortised Inference
• Reinforcement learning with world model
• Worrisome Properties of Neural Network Controllers and Their Symbolic Representations
• An Entropy Regularization Free Mechanism for Policy-based Reinforcement Learning
• Metis: Multi-Agent Based Crisis Simulation System
• Lyapunov-Based Reinforcement Learning for Decentralized Multi-Agent Control
• PFPN: Continuous Control of Physically Simulated Characters using Particle Filtering Policy Network
• Continuous Control With Ensemble Deep Deterministic Policy Gradients
• Learning Policies through Quantile Regression
• Improving the Exploration of Deep Reinforcement Learning in Continuous Domains using Planning for Policy Search
• Sample Efficient Social Navigation Using Inverse Reinforcement Learning
• Planning and Learning Using Adaptive Entropy Tree Search
• Decoupled Exploration and Exploitation Policies for Sample-Efficient Reinforcement Learning
• Co-Adaptation of Algorithmic and Implementational Innovations in Inference-based Deep Reinforcement Learning
• Model-free Policy Learning with Reward Gradients
• Model Predictive Actor-Critic: Accelerating Robot Skill Acquisition with Deep Reinforcement Learning
• Utilizing Skipped Frames in Action Repeats via Pseudo-Actions
• Learning of feature points without additional supervision improves reinforcement learning from images
• Towards Learning to Play Piano with Dexterous Hands and Touch
• Stateless actor-critic for instance segmentation with high-level priors
• Optimizing Quantum Variational Circuits with Deep Reinforcement Learning
• Deep Reinforcement Learning Based Networked Control with Network Delays for Signal Temporal Logic Specifications
• Count-Based Temperature Scheduling for Maximum Entropy Reinforcement Learning
• Homotopy Based Reinforcement Learning with Maximum Entropy for Autonomous Air Combat
• Learning Transferable Concepts in Deep Reinforcement Learning
• Prioritized Soft Q-Decomposition for Lexicographic Reinforcement Learning
• Optimal operation of cryogenic calorimeters through deep reinforcement learning
• Predictable Interval MDPs through Entropy Regularization
• Enhancing Reinforcement Learning Through Guided Search
• Distributed Beamforming in Massive MIMO Communication for a Constellation of Airborne Platform Stations
• Variational Policy Search using Sparse Gaussian Process Priors for Learning Multimodal Optimal Actions
• Adaptation of Quadruped Robot Locomotion with Meta-Learning
• A Model-free Deep Reinforcement Learning Approach To Maneuver A Quadrotor Despite Single Rotor Failure
• Efficiently Training On-Policy Actor-Critic Networks in Robotic Deep Reinforcement Learning with Demonstration-like Sampled Exploration
• OffCon${}^3$: What is state of the art anyway?
• Model-Augmented Q-learning
• Maximum Entropy Reinforcement Learning with Mixture Policies
• A Bayesian Approach to Reinforcement Learning of Vision-Based Vehicular Control
• Learning to Reweight Imaginary Transitions for Model-Based Reinforcement Learning
• Developing cooperative policies for multi-stage tasks
• Deep Reinforcement Learning for Tactile Robotics: Learning to Type on a Braille Keyboard
• Biomechanic Posture Stabilisation via Iterative Training of Multi-policy Deep Reinforcement Learning Agents
• Predicting Sim-to-Real Transfer with Probabilistic Dynamics Models
• RLCache: Automated Cache Management Using Reinforcement Learning