Theseus: A Library for Differentiable Nonlinear Optimization
arXiv:2207.09442
Abstract
We present Theseus, an efficient application-agnostic open source library for differentiable nonlinear least squares (DNLS) optimization built on PyTorch, providing a common framework for end-to-end structured learning in robotics and vision. Existing DNLS implementations are application specific and do not always incorporate many ingredients important for efficiency. Theseus is application-agnostic, as we illustrate with several example applications that are built using the same underlying differentiable components, such as second-order optimizers, standard costs functions, and Lie groups. For efficiency, Theseus incorporates support for sparse solvers, automatic vectorization, batching, GPU acceleration, and gradient computation with implicit differentiation and direct loss minimization. We do extensive performance evaluation in a set of applications, demonstrating significant efficiency gains and better scalability when these features are incorporated. Project page: https://sites.google.com/view/theseus-ai
Advances in Neural Information Processing Systems (NeurIPS), 2022
Cited by in corpus (8)
- A Survey of Contextual Optimization Methods for Decision Making under Uncertainty
- Autonomous Drone Racing: A Survey
- Digital Twins in Wind Energy: Emerging Technologies and Industry-Informed Future Directions
- PyPose: A Library for Robot Learning with Physics-based Optimization
- Accelerating Optimal Power Flow with GPUs: SIMD Abstraction of Nonlinear Programs and Condensed-Space Interior-Point Methods
- Drivable Avatar Clothing: Faithful Full-Body Telepresence with Dynamic Clothing Driven by Sparse RGB-D Input
- Leveraging Cycle-Consistent Anchor Points for Self-Supervised RGB-D Registration
- Uncertainty Quantification of Set-Membership Estimation in Control and Perception: Revisiting the Minimum Enclosing Ellipsoid