(2022). SMODICE: Versatile Offline Imitation Learning via State Occupancy Matching. In ICML.

Cite arXiv Webpage Code

(2022). Fighting Fire with Fire: Avoiding DNN Shortcuts through Priming. ICML.


(2022). Know Thyself: Transferable Visuomotor Control Through Robot-Awareness. In ICLR.

Cite arXiv Webpage

(2022). Prospective Learning: Back to the Future. In arXiv.

Cite arXiv

(2022). Conservative and Adaptive Penalty for Model-Based Safe Reinforcement Learning. AAAI.

Cite arXiv Code

(2021). Keyframe-Focused Visual Imitation Learning. In ICML.

Cite arXiv Webpage

(2021). Femtomolar SARS-CoV-2 Antigen Detection Using the Microbubbling Digital Assay with Smartphone Readout Enables Antigen Burden Quantitation and Dynamics Tracking. Clinical Chemistry.

Cite medRxiv

(2021). Diverse Sampling for Normalizing Flow Based Trajectory Forecasting. ICCV.

Cite arXiv

(2021). Conservative Offline Distributional Reinforcement Learning. In NeurIPS.

Cite arXiv Code

(2021). An Exploration of Embodied Visual Exploration. In IJCV.

Cite arXiv Webpage IJCV version

(2021). SMIRL: Surprise Minimizing RL in Dynamic Environments. In ICLR.

Cite arXiv Webpage

(2021). Object Representations Guided By Optical Flow. NeurIPS Robot Learning Workshop.


(2021). Embracing the Reconstruction Uncertainty in 3D Human Pose Estimation. ICCV.


(2020). Model-Based Inverse Reinforcement Learning from Visual Demonstrations. In CORL.

Cite arXiv Webpage

(2020). Fighting Copycat Agents in Behavioral Cloning from Multiple Observations.. In NeurIPS.

Cite arXiv

(2020). Long-Horizon Visual Planning with Goal-Conditioned Hierarchical Predictors. In NeurIPS.

PDF Cite arXiv Webpage Video

(2020). Cautious Adaptation For Reinforcement Learning in Safety-Critical Settings. ICML.

Cite arXiv ICML Proceedings Webpage Code

(2020). MAVRIC: Morphology-Agnostic Visual Robotic Control. In ICRA and RA-L.

PDF Cite arXiv Webpage

(2020). DIGIT: A Novel Design for a Low-Cost Compact High-Resolution Tactile Sensor with Application to In-Hand Manipulation. In ICRA and RA-L.

PDF Cite RA-L Webpage

(2019). Causal Confusion in Imitation Learning. NeurIPS (Oral).

PDF Cite arXiv Webpage

(2019). Time-Agnostic Prediction: Predicting Predictable Video Frames. In ICLR.

PDF Cite arXiv Webpage

(2019). REPLAB: A Reproducible Low-Cost Arm Benchmark Platform for Robotic Learning. In ICRA.

PDF Cite arXiv Webpage

(2019). REPLAB: A Reproducible Low-Cost Arm Benchmark Platform for Robotic Learning. arXiv preprint (extended version of earlier ICRA publication).

PDF Cite arXiv Webpage

(2019). Manipulation by Feel: Touch-Based Control with Deep Predictive Models. In ICRA.

PDF Cite arXiv BAIR blog post Webpage

(2019). Emergence of Exploratory Look-Around Behaviors Through Active Observation Completion. In Science Robotics.

PDF Cite Online article Webpage

(2018). Techniques for Rectification of Camera Arrays.

PDF Cite Google Patents

(2018). ShapeCodes: Self-Supervised Feature Learning by Lifting Views to Viewgrids. In ECCV.

PDF Cite arXiv

(2018). More Than a Feeling: Learning to Grasp and Regrasp using Vision and Touch. In IROS and RAL (RAL Best Paper Runner-Up).

PDF Cite arXiv Webpage

(2018). End-to-End Policy Learning For Active Visual Categorization. In IEEE TPAMI.

PDF Cite Preprint

(2017). Techniques for Improved Focusing of Camera Arrays.

PDF Cite Google Patents

(2017). Learning Image Representations Tied to Egomotion from Unlabeled Video. In IJCV Special Issue of Best Papers from ICCV 2015.

PDF Cite Downloadable Preprint Webpage

(2017). Embodied Learning for Visual Recognition. The University of Texas at Austin.

PDF Cite

(2017). Divide, Share, and Conquer: Multi-Task Attribute Learning With Selective Sharing. In Springer Book on Visual Attributes.

PDF Cite Book

(2016). Pano2Vid: Automatic Cinematography For Watching 360-degree Videos. In ACCV (Songde Ma Best Application Paper Award).

PDF Cite Supplementary pdf Webpage arXiv

(2016). Object-Centric Representation Learning from Unlabeled Videos. In ACCV.

PDF Cite arXiv Webpage

(2012). Objective Quality Assessment of Multiply Distorted Images. In ASILOMAR Conference on Signals, Systems and Computers.

PDF Cite Webpage