Cooper: A Library for Constrained Optimization in Deep Learning
This provides a tool for researchers and practitioners in deep learning to handle constrained optimization problems, but it is incremental as it builds on existing Lagrangian methods.
The authors introduced Cooper, an open-source library for constrained optimization in deep learning, which implements Lagrangian-based first-order update schemes to integrate with PyTorch features like automatic differentiation and specialized architectures.
Cooper is an open-source package for solving constrained optimization problems involving deep learning models. Cooper implements several Lagrangian-based first-order update schemes, making it easy to combine constrained optimization algorithms with high-level features of PyTorch such as automatic differentiation, and specialized deep learning architectures and optimizers. Although Cooper is specifically designed for deep learning applications where gradients are estimated based on mini-batches, it is suitable for general non-convex continuous constrained optimization. Cooper's source code is available at https://github.com/cooper-org/cooper.