Differentiable Spline Approximations

Abstract

The paradigm of differentiable programming has significantly enhanced the scope of machine learning via the judicious use of gradient-based optimization. However, standard differentiable programming methods (such as autodiff) typically require that the machine learning models be differentiable, limiting their applicability. Our goal in this paper is to use a new, principled approach to extend gradient-based optimization to functions well modeled by splines, which encompass a large family of piecewise polynomial models. We derive the form of the (weak) Jacobian of such functions and show that it exhibits a block-sparse structure that can be computed implicitly and efficiently. Overall, we show that leveraging this redesigned Jacobian in the form of a differentiable "layer" in predictive models leads to improved performance in diverse applications such as image segmentation, 3D point cloud reconstruction, and finite element analysis.

9 pages, accepted in Neurips 2021

References in corpus (8)

• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Differentiable Convex Optimization Layers
• DiffTaichi: Differentiable Programming for Physical Simulation
• Physics-informed neural networks (PINNs) for fluid mechanics: A review
• Learning to Control PDEs with Differentiable Physics
• Scalable Differentiable Physics for Learning and Control
• Accelerating PDE-constrained Inverse Solutions with Deep Learning and Reduced Order Models
• Differentiable Top-k Operator with Optimal Transport