News

Grasp2Grasp: Vision-Based Dexterous Grasp Translation via Schrödinger Bridges accepted at NeurIPS 2025
Nov. 10, 2025

We propose a new approach to vision-based dexterous grasp translation, which aims to transfer grasp intent across robotic hands with differing morphologies. Given a visual observation of a source hand grasping an object, our goal is to synthesize a functionally equivalent grasp for a target hand without requiring paired demonstrations or hand…

Geometric Algebra Grasp Diffusion for Dexterous Manipulators accepted at ICRA 2025
Jan. 27, 2025

We propose a novel framework for dexterous grasp generation that leverages geometric algebra representations to enforce equivariance to SE(3) transformations. By encoding the SE(3) symmetry constraint directly into the architecture, our method improves data and parameter efficiency, while enabling robust grasp generation across diverse object…

Learning Color Equivariant Representations accepted at ICLR 2025
Jan. 22, 2025

In this paper, we introduce group convolutional neural networks (GCNNs) equivariant to color variation. GCNNs have been designed for a variety of geometric transformations from 2D and 3D rotation groups, to semi-groups such as scale. Despite the improved interpretability, accuracy and generalizability of these architectures, GCNNs have seen…

Behavior-Inspired Neural Network for Relational Reasoning accepted at AISTATS 2025
Jan. 21, 2025

From pedestrians to Kuramoto oscillators, interactions between agents govern how a multitude of dynamical systems evolve in space and time. Discovering how these agents relate to each other can improve our understanding of the often complex dynamics that underlie these systems. Recent works learn to categorize relationships between agents based…

Geometric Algebra Grasp Diffusion for Dexterous Manipulators presented at IROS 2024 Workshop Equivariant Robotics
Oct. 14, 2024

We propose a novel framework for dexterous grasp generation that leverages geometric algebra representations to enforce equivariance to SE(3) transformations. By encoding the SE(3) symmetry constraint directly into the architecture, our method improves data and parameter efficiency, while enabling robust grasp generation across diverse object…

Color Equivariant Network presented at CVPR 2024 Workshop Equivariant Vision
June 18, 2024

Group equivariant convolutional neural networks have been designed for a variety of geometric transformations from 2D and 3D rotation groups, to semi-groups such as scale. Despite the improved interpretability, accuracy and generalizability afforded by these architectures, group equivariant networks have seen limited application in the context…

UCSD controls seminar speaker
April 12, 2024

Learning Hamiltonian Dynamics from Video

Keynote at the New York University, NYC Computer Vision Day
Nov. 4, 2023

The NYC Computer Vision Day is an invite-only event that aims to be an informal day where the computer vision community from NYC and surroundings can share ideas and meet. A primary focus is visibility for graduate students and early career researchers. In addition to a strong showing from ≈ 260 researchers from 60+ research labs and 15+…

Hamiltonian GAN accepted at L4DC 2024
Sept. 27, 2023

A growing body of work leverages the Hamiltonian formalism as an inductive bias for physically plausible neural network based video generation. The structure of the Hamiltonian ensures conservation of a learned quantity (e.g., energy) and imposes a phase-space interpretation on the low-dimensional manifold underlying the input video. While this…