NNGeometry: Easy and Fast Fisher Information Matrices and Neural Tangent Kernels in PyTorch

Thomas George

Abstract: Fisher Information Matrices (FIM) and Neural Tangent Kernels (NTK) are useful tools in a number of diverse applications related to neural networks. Yet these theoretical tools are often difficult to implement using current libraries for practical size networks, given that they require per-example gradients, and a large amount of memory since they scale as the number of parameters (for the FIM) or the number of examples x cardinality of the output space (for the NTK). NNGeometry is a PyTorch library that offers a simple interface for computing various linear algebra operations such as matrix-vector products, trace, frobenius norm, and so on, where the matrix is either the FIM or the NTK, leveraging recent advances in approximating these matrices. We here present the library and motivate our design choices, then we demonstrate it on actual deep neural networks.

One-sentence Summary: A PyTorch library for computing Fisher Information Matrices and Neural Tangent Kernels

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=pxpkEve6_j