Latest

• Discovering Robotic Interaction Modes with Discrete Representation Learning
• SPIRE: Synergistic Planning, Imitation, and Reinforcement for Long-Horizon Manipulation
• Fast Explicit-Input Assistance for Teleoperation in Clutter
• SuFIA: Language-Guided Augmented Dexterity for Robotic Surgical Assistants
• Adaptive Horizon Actor-Critic for Policy Learning in Contact-Rich Differentiable Simulation
• HandyPriors: Physically Consistent Perception of Hand-Object Interactions with Differentiable Priors
• ORBIT-Surgical: An Open-Simulation Framework for Learning Surgical Augmented Dexterity
• DexGrasp-1M: Dexterous Multi-finger Grasp Generation Through Differentiable Simulation
• MVTrans: Multi-View Perception of Transparent Objects
• nerf2nerf: Pairwise Registration of Neural Radiance Fields
• ProgPrompt: Generating Situated Robot Task Plans using Large Language Models
• Self-Supervised Learning of Action Affordances as Interaction Modes
• Learning Achievement Structure for Structured Exploration in Domains with Sparse Reward
• SlotFormer: Unsupervised Visual Dynamics Simulation with Object-Centric Models
• ORBIT : A Unified Simulation Framework for Interactive Robot Learning Environments
• Bayesian Object Models for Robotic Interaction with Differentiable Probabilistic Programming
• Breaking Bad: A Dataset for Geometric Fracture and Reassembly
• MoCoDA: Model-based Counterfactual Data Augmentation
• RoboTube: Learning Household Manipulation from Human Videos with Simulated Twin Environments
• SMPL: Simulated Industrial Manufacturing and Process Control Learning Environments
• Articulated Object Interaction in Unknown Scenes with Whole-Body Mobile Manipulation.
• Grasp'D: Differentiable Contact-rich Grasp Synthesis for Multi-fingered Hands
• Transferring Dexterous Manipulation from GPU Simulation to a Remote Real-World TriFinger
• DiSECt: A Differentiable Simulator for Parameter Inference and Control in Robotic Cutting
• Koopman Q-learning: Offline Reinforcement Learning via Symmetries of Dynamics
• GLiDE: Generalizable Quadrupedal Locomotion in Diverse Environments with a Centroidal Model
• Modular Action Concept Grounding in Semantic Video Prediction
• Neural Shape Mating: Self-Supervised Object Assembly with Adversarial Shape Priors
• PlaTe: Visually-Grounded Planning with Transformers in Procedural Tasks
• Accelerated Policy Learning with Parallel Differentiable Simulation
• Pessimistic Bootstrapping for Uncertainty-Driven Offline Reinforcement Learning
• Value Gradient weighted Model-Based Reinforcement Learning
• Drop-DTW: Aligning Common Signal Between Sequences While Dropping Outliers
• Dynamic Bottleneck for Robust Self-Supervised Exploration
• Neural Hybrid Automata: Learning Dynamics With Multiple Modes and Stochastic Transitions
• A Persistent Spatial Semantic Representation for High-level Natural Language Instruction Execution
• S4RL: Surprisingly Simple Self-Supervision for Offline Reinforcement Learning in Robotics
• Seeing Glass: Joint Point-Cloud and Depth Completion for Transparent Objects
• Uniform Priors for Data-Efficient Transfer
• Learning by Watching: Physical Imitation of Manipulation Skills from Human Videos
• A differentiable simulator for robotic cutting
• Coach-Player Multi-agent Reinforcement Learning for Dynamic Team Composition.
• DiSeCT: A differentiable simulation engine for autonomous robotic cutting
• GIFT: Generalizable Interaction-aware Functional Tool Affordances without Labels.
• Principled Exploration via Optimistic Bootstrapping and Backward Induction.
• Tesseract: Tensorised Actors for Multi-Agent Reinforcement Learning
• Dynamics Randomization Revisited: A Case Study for Quadrupedal Locomotion
• Emergent Hand Morphology and Control from Optimizing Robust Grasps of Diverse Objects
• LASER: Learning a Latent Action Space for Efficient Reinforcement Learning
• LEAF: Latent Exploration Along the Frontier.
• C-Learning: Horizon-Aware Cumulative Accessibility Estimation
• Conservative Safety Critics for Exploration
• Skill Transfer via Partially Amortized Hierarchical Planning
• DIBS: Diversity inducing Information Bottleneck in Model Ensembles
• Unsupervised Disentanglement of Pose, Appearance and Background from Images and Videos
• Causal Discovery in Physical Systems from Videos.
• Counterfactual Data Augmentation using Locally Factored Dynamics
• Curriculum By Smoothing
• D2RL: Deep Dense Architectures in Reinforcement Learning
• Emergent Hand Morphology and Control from Optimizing Robust Grasps of Diverse Objects
• Learning a Contact-Adaptive Controller for Robust, Efficient Legged Locomotion
• Visuomotor Mechanical Search: Learning to Retrieve Target Objects in Clutter
• Ocean: Online Task Inference for Compositional Tasks with Context Adaptation
• A Programmable Approach To Model Compression
• Angular Visual Hardness
• Semi-Supervised StyleGAN for Disentanglement Learning
• Controlling Assistive Robots with Learned Latent Actions
• Guided Uncertainty-Aware Policy Optimization: Combining Model-Free and Model-Based Strategies for Sample-Efficient Learning
• Implicit Reinforcement without Interaction at Scale: Leveraging Large-Scale Robot Manipulation Datasets for Control
• Motion Reasoning for Goal-Based Imitation Learning
• Combining Model-Free and Model-Based Strategies for Sample-Efficient Reinforcement Learning
• InfoCNF: An Efficient Conditional Continuous Normalizing Flow with Adaptive Solvers
• Continuous Relaxation of Symbolic Planner for One-Shot Imitation Learning
• Scaling Robot Supervision to Hundreds of Hours with RoboTurk: Robotic Manipulation Dataset through Human Reasoning and Dexterity
• Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Tasks
• AC-Teach: A Bayesian Actor-Critic Method for Policy Learning with an Ensemble of Suboptimal Teachers
• Dynamics Learning with Cascaded Variational Inference for Multi-Step Manipulation
• Video Interpolation and Prediction with Unsupervised Landmarks