[go: up one dir, main page]

Lerrel Pinto

I am the Co-founder and CEO of Assured Robot Intelligence (ARI). We are building frontier robots to make physical labor abundant for everyone. If you are interested in joining our mission, please send us an email.

Research: My research focus is on large-scale robot learning (both data and models), sensory representation learning, action and behavior modeling, reinforcement learning for adaptation, and the development of affordable, open-source robots. See recent papers here

Recent News

Recent Talks

Here are some public talks that covers my recent research:

Courses Taught at NYU

Selected Research and Publications


GIF
Robot Utility Models: General Policies for Zero-Shot Deployment in New Environments

GIF
AnySkin: Plug-and-play Skin Sensing for Robotic Touch

GIF
Bridging the Human to Robot Dexterity Gap through Object-Oriented Rewards

GIF
DynaMem: Online Dynamic Spatio-Semantic Memory for Open World Mobile Manipulation

GIF
P3-PO: Prescriptive Point Priors for Visuo-Spatial Generalization of Robot Policies


GIF
DynaMo: In-Domain Dynamics Pretraining for Visuo-Motor Control

GIF
BAKU: An Efficient Transformer for Multi-Task Policy Learning

GIF
OPEN TEACH: A Versatile Teleoperation System for Robotic Manipulation

GIF
OK-Robot: What Really Matters in Integrating Open-Knowledge Models for Robotics

GIF
VQ-BeT: Behavior Generation with Latent Actions

GIF
Adaptive Sampling of k-space in Magnetic Resonance for Rapid Pathology Prediction

GIF
Hierarchical State Space Models for Continuous Sequence-to-Sequence Modeling

GIF
diff History for Neural Language Agents

GIF
On Bringing Robots Home

GIF
See to Touch: Learning Tactile Dexterity through Visual Incentives

GIF
NetHack is Hard to Hack



GIF
Teach a Robot to FISH: Versatile Imitation from One Minute of Demonstrations

GIF
CLIP-Fields: Weakly Supervised Semantic Fields for Robotic Memory

GIF
From play to policy: Conditional behavior generation from uncurated robot data

GIF
Holo-Dex: Teaching Dexterity with Immersive Mixed Reality

GIF
Behavior Transformers: Cloning k modes with one stone

GIF
Watch and Match: Supercharging Imitation with Regularized Optimal Transport

GIF
Dexterous Imitation Made Easy: A Learning-Based Framework for Efficient Dexterous Manipulation

GIF
Don't Change the Algorithm, Change the Data: Exploratory Data for Offline Reinforcement Learning

GIF
The Surprising Effectiveness of Representation Learning for Visual Imitation


GIF
Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learning

GIF
Reinforcement Learning with Prototypical Representations

GIF
Learning Cross-Domain Correspondence for Control with Dynamics Cycle-Consistency

GIF
Self-Supervised Policy Adaptation during Deployment

GIF
Task-Agnostic Morphology Evolution

GIF
Visual Imitation Made Easy

GIF
Learning Predictive Representations for Deformable Objects Using Contrastive Estimation

GIF
Robust Policies via Mid-Level Visual Representations

GIF
Automatic Curriculum Learning through Value Disagreement

GIF
Generalized Hindsight for Reinforcement Learning

GIF
Reinforcement Learning with Augmented Data

GIF
Learning to Manipulate Deformable Objects without Demonstrations

GIF
Swoosh! Rattle! Thump! - Actions that Sound

GIF
Hierarchically Decoupled Imitation for Morphological Transfer

GIF
PyRobot: An Open-source Robotics Framework for Research and Benchmarking

GIF
Multiple Interactions Made Easy (MIME): Large Scale Demonstrations Data for Imitation

GIF
Robot Learning in Homes: Improving Generalization and Reducing Dataset Bias

GIF
Asymmetric Actor Critic for Image-Based Robot Learning

GIF
CASSL: Curriculum Accelerated Self-Supervised Learning

GIF
Learning to Fly by Crashing

GIF
Robust Adversarial Reinforcement Learning


GIF
Supervision via Competition: Robot Adversaries for Learning Tasks

GIF
The Curious Robot: Learning Visual Representations via Physical Interactions

GIF
Supersizing Self-supervision: Learning to Grasp from 50K Tries and 700 Robot Hours