TorchOpt: An Efficient Library for Differentiable Optimization
University of Edinburgh, University of Edinburgh · University College London · National University of Singapore · Peking University · University of Edinburgh
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
Differentiable optimization algorithms often involve expensive computations of various meta-gradients. To address this, we design and implement TorchOpt, a new PyTorch-based differentiable optimization library. TorchOpt provides an expressive and unified programming interface that simplifies the implementation of explicit, implicit, and zero-order gradients. Moreover, TorchOpt has a distributed execution runtime capable of parallelizing diverse operations linked to differentiable optimization tasks across CPU and GPU devices. Experimental results demonstrate that TorchOpt achieves a 5.2× training time speedup in a cluster. TorchOpt is open-sourced at https://github.com/metaopt/torchopt and has become a PyTorch Ecosystem project.