Self-Supervised Learning

Discovering Deformable Keypoint Pyramids
Jianing Qian, Anastasios Panagopoulos, Dinesh Jayaraman
An Exploration of Embodied Visual Exploration
Santhosh K Ramakrishnan*, Dinesh Jayaraman, Kristen Grauman
An Exploration of Embodied Visual Exploration
Model-Based Inverse Reinforcement Learning from Visual Demonstrations
We learn reward functions in unsupervised object keypoint space, to allow us to follow third-person demonstrations with model-based RL.
Neha Das, Sarah Bechtle, Todor Davchev, Dinesh Jayaraman, Akshara Rai, Franziska Meier
DIGIT: A Novel Design for a Low-Cost Compact High-Resolution Tactile Sensor with Application to In-Hand Manipulation
We design and demonstrate a new tactile sensor for in-hand tactile manipulation in a robotic hand.
Mike Lambeta, Po-Wei Chou, Stephen Tian, Brian Yang, Benjamin Maloon, Victoria Rose Most, Dave Stroud, Raymond Santos, Ahmad Byagowi, Gregg Kammerer, Dinesh Jayaraman, Roberto Calandra
DIGIT: A Novel Design for a Low-Cost Compact High-Resolution Tactile Sensor with Application to In-Hand Manipulation
MAVRIC: Morphology-Agnostic Visual Robotic Control
We demonstrate visual control within 20 seconds on a robot with unknown morphology, from a single uncalibrated RGBD camera.
Brian Yang, Dinesh Jayaraman, Glen Berseth, Alexei Efros, Sergey Levine
MAVRIC: Morphology-Agnostic Visual Robotic Control
Emergence of Exploratory Look-Around Behaviors Through Active Observation Completion
Santhosh K Ramakrishnan*, Dinesh Jayaraman, Kristen Grauman
Manipulation by Feel: Touch-Based Control with Deep Predictive Models
High-resolution tactile sensing together with visual approaches to prediction and planning with deep neural networks enables high-precision tactile servoing tasks.
Stephen Tian, Frederik Ebert, Dinesh Jayaraman, Mayur Mudigonda, Chelsea Finn, Roberto Calandra, Sergey Levine
Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks
Task-agnostic visual exploration policies may be trained through a proxy “observation completion” task that requires an agent to “paint” unobserved views given a small set of observed views.
Dinesh Jayaraman, Kristen Grauman
More Than a Feeling: Learning to Grasp and Regrasp using Vision and Touch
By exploiting high precision tactile sensing with deep learning, robots can effectively iteratively adjust their grasp configurations to boost grasping performance from 65% to 94%.
Roberto Calandra, Andrew Owens, Dinesh Jayaraman, Justin Lin, Wenzhen Yuan, Jitendra Malik, Edward H Adelson, Sergey Levine
ShapeCodes: Self-Supervised Feature Learning by Lifting Views to Viewgrids
Appearance-based image representations in the form of viewgrids provide a useful framework for learning self-supervised image representations by training a network to reconstruct full object shapes, or scenes.
Dinesh Jayaraman, Ruohan Gao, Kristen Grauman