A curated list of papers of interesting empirical study and insight on deep learning. Continually updating...

{KFAC,EKFAC,Diagonal,Implicit} Fisher Matrices and finite width NTKs in PyTorch

A fast, effective data attribution method for neural networks in PyTorch

Multi-framework implementation of Deep Kernel Shaping and Tailored Activation Transformations, which are methods that modify neural network models (and their initializations) to make them easier to train.

codebase for "A Theory of the Inductive Bias and Generalization of Kernel Regression and Wide Neural Networks"

Neural Tangent Kernel (NTK) module for the scikit-learn library

We propose a lossless compression algorithm based on the NTK matrix for DNN. The compressed network yields asymptotically the same NTK as the original (dense and unquantized) network, with its weights and activations taking values only in {0, 1, -1} up to scaling.

Official repository of our work "Finding Lottery Tickets in Vision Models via Data-driven Spectral Foresight Pruning" accepted at CVPR 2024

Study of the paper 'Neural Thompson Sampling' published in October 2020

ICML2025-Inductive Gradient Adjustment for Spectral Bias in Implicit Neural Representations

Official Code: Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes

Code accompanying the paper "On the adaptation of recurrent neural networks for system identification"

TF2 Implementation of Physics Informed Neural Networks and Neural Tangent Kernel

A Python implementation of the Neural Tangent Kernel (jacot et al, 2018)

Code for "Learnware Reduced Kernel Mean Embedding Specification Based on Neural Tangent Kernel"

Implementation of "Deep Learning in Random Neural Fields: Numerical Experiments via Neural Tangent Kernel"

TCT: Convexifying Federated Learning using Bootstrapped Neural Tangent Kernels

Understand the spectral bias of deep learning through the study of NTK

Senior Project for Statistics & Data Science at Yale University

This project explores fault detection in ball bearings using Support Vector Machines (SVM) with Neural Tangent Kernel (NTK). By leveraging advanced machine learning techniques on vibration signal data, we achieve high-accuracy predictive maintenance, helping to prevent machine failures and optimize industrial operations.