No article text was recorded for this version.
Kronecker-factored curvature approximation
Version 2 of 2239 papers313 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Fisher Information and Natural Gradient Learning in Random Deep Networks
- Blog: Kronecker-Factored Approximate Curvature for Physics-Informed Neural Networks
- Blog: A New Perspective on Shampoo's Preconditioner
- Blog: AdaFisher: Adaptive Second Order Optimization via Fisher Information
- Blog: An Evaluation of Fisher Approximations Beyond Kronecker Factorization
- Blog: SKFAC: Training Neural Networks With Faster Kronecker-Factored Approximate Curvature
- Blog: Fast Approximate Natural Gradient Descent in a Kronecker-factored Eigenbasis
- Blog: Efficient Bilevel Optimization with KFAC-Based Hypergradients
- Blog: Optimizing Neural Networks with Kronecker-factored Approximate Curvature
- Blog: Exact natural gradient in deep linear networks and its application to the nonlinear case
- Blog: Enhancing Prediction Performance through Influence Measure
- Blog: Practical Quasi-Newton Methods for Training Deep Neural Networks
- Blog: FedBEns: One-Shot Federated Learning based on Bayesian Ensemble
- Blog: Layer-Wise Conditioning Analysis in Exploring the Learning Dynamics of DNNs
- Blog: Bag of Tricks for Natural Policy Gradient Reinforcement Learning
- Blog: AdaFisher: Adaptive Second Order Optimization via Fisher Information
- Blog: Scalable Sparse Covariance Estimation via Self-Concordance
- Blog: A Unified Theory of Random Projection for Influence Functions
- Blog: Eva: Practical Second-order Optimization with Kronecker-vectorized Approximation
- Blog: Fishy: Layerwise Fisher Approximation for Higher-order Neural Network Optimization
- Blog: Fast Approximate Natural Gradient Descent in a Kronecker Factored Eigenbasis
- Blog: Second-order Optimization Method for Large Mini-batch: Training ResNet-50 on ImageNet in 35 Epochs
- Blog: Eigenvalue-corrected Natural Gradient Based on a New Approximation
- Blog: Online Structured Laplace Approximations for Overcoming Catastrophic Forgetting
- Blog: BackPACK: Packing more into Backprop
- Blog: An Overview of Stochastic Quasi-Newton Methods for Large-Scale Machine Learning
- Blog: PredProp: Bidirectional Stochastic Optimization with Precision Weighted Predictive Coding
- Blog: DyKAF: Dynamical Kronecker Approximation of the Fisher Information Matrix for Gradient Preconditioning
- Blog: TENGraD: Time-Efficient Natural Gradient Descent with Exact Fisher-Block Inversion
- Blog: Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation
- Blog: Enhance Curvature Information by Structured Stochastic Quasi-Newton Methods
- Blog: A Novel Structured Natural Gradient Descent for Deep Learning
- Blog: New Insights and Perspectives on the Natural Gradient Method
- Blog: A Scalable Laplace Approximation for Neural Networks
- Blog: A Kronecker-factored approximate Fisher matrix for convolution layers
- Blog: Kronecker-Factored Approximate Curvature for Physics-Informed Neural Networks
- Blog: Kronecker-factored Approximate Curvature (KFAC) From Scratch
- Blog: Low-Rank Curvature for Zeroth-Order Optimization in LLM Fine-Tuning
- Blog: Stochastic Gradient Langevin dynamics that Exploit Neural Network Structure
- Blog: Scalable and Practical Natural Gradient for Large-Scale Deep Learning
- Blog: Towards Efficient Optimizer Design for LLM via Structured Fisher Approximation with a Low-Rank Extension
- Blog: Structured Inverse-Free Natural Gradient Descent: Memory-Efficient & Numerically-Stable KFAC
- Blog: Efficiently Access Diffusion Fisher: Within the Outer Product Span Space
- Blog: Natural Policy Gradient Methods with Parameter-based Exploration for Control Tasks
- Blog: Scalable Multi-Stage Influence Function for Large Language Models via Eigenvalue-Corrected Kronecker-Factored Parameterization
- Blog: Towards Practical Second-Order Optimizers in Deep Learning: Insights from Fisher Information Analysis
- Blog: FedFisher: Leveraging Fisher Information for One-Shot Federated Learning
- Blog: Dissecting Hessian: Understanding Common Structure of Hessian in Neural Networks
- Blog: Kronecker-factored Curvature Approximations for Recurrent Neural Networks
- Blog: Feedforward and Recurrent Neural Networks Backward Propagation and Hessian in Matrix Form
- Blog: Online Curvature-Aware Replay: Leveraging 2nd Order Information for Online Continual Learning
- Blog: Scalable Stochastic Gradient Riemannian Langevin Dynamics in Non-Diagonal Metrics
- Blog: Distributed Second-Order Optimization using Kronecker-Factored Approximations
- Blog: BackPACK: Packing more into backprop
- Blog: Invariance Learning in Deep Neural Networks with Differentiable Laplace Approximations
- Blog: Scalable Thermodynamic Second-order Optimization
- Blog: A Trace-restricted Kronecker-Factored Approximation to Natural Gradient
- Blog: Understanding Approximate Fisher Information for Fast Convergence of Natural Gradient Descent in Wide Neural Networks
- Blog: Riemannian Natural Gradient Methods
- Blog: A Mini-Block Fisher Method for Deep Neural Networks
Showing 60 of 239 papers.
History & citation
Current · Permanent link
Cite this direction
Latest version (a living page):
@misc{lacuna_kronecker_factored_curvature_approximati_a50872ce,
title = {Kronecker-factored curvature approximation},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/kronecker-factored-curvature-approximation/rdr_a50872cef632437aa307aaddf116988c},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}This exact version (version 2, frozen text):
@misc{lacuna_kronecker_factored_curvature_approximati_a50872ce_v2,
title = {Kronecker-factored curvature approximation},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/kronecker-factored-curvature-approximation/rdr_a50872cef632437aa307aaddf116988c/v/2},
note = {Version 2, frozen 2026-10-04. Accessed 2026-11-20},
}