Ricky Chen

Papers

Total Citations

H-Index

About

Ricky Chen is a leading researcher at the intersection of robotics, computer vision, and differentiable optimization, whose work is reshaping how machines learn from structured data. Chen's primary contributions center on making nonlinear optimization differentiable and accessible, enabling end-to-end learning pipelines that were previously fragmented across specialized domains. Their landmark work, "Theseus: A Library for Differentiable Nonlinear Optimization" (2022), has already garnered 47 citations, establishing a unified, open-source framework built on PyTorch that allows researchers to seamlessly integrate nonlinear least squares optimization into deep learning architectures. This innovation has profound implications for robotics and vision tasks—from visual odometry to motion planning—where traditional optimization and modern learning were historically siloed. Chen's approach democratizes complex optimization, allowing students and practitioners to build end-to-end systems without reinventing the wheel. By bridging the gap between classical robotics and contemporary deep learning, Chen is not only advancing autonomous systems but also creating tools that empower the next generation of researchers to tackle real-world perception and control challenges with unprecedented efficiency and elegance.