The is the official implementation of ICML 2023 paper "Revisiting Weighted Aggregation in Federated Learning with Neural Networks".
-
Updated
Aug 28, 2023 - Python
#
weight-decay
Here are 14 public repositories matching this topic...
[NeurIPS 2023] The PyTorch Implementation of Scheduled (Stable) Weight Decay.
-
Updated
Feb 3, 2024 - Python
AdamW optimizer engine decoupling L2 weight decay regularization from adaptive gradient first and second moments.
-
Updated
Sep 10, 2026 - Python
AdamW optimizer engine decoupling L2 weight decay regularization from adaptive gradient first and second moments.
-
Updated
Sep 10, 2026 - Python
Investigating the Role of Weight Decay in Enhancing Nonconvex SGD, CVPR 2025
-
Updated
Jun 9, 2025 - Python
Advanced CIFAR-10 image classification using ResNet-inspired CNN with residual blocks, achieving 92%+ accuracy through comprehensive regularization, data augmentation, and professional ML engineering practices.
computer-vision deep-learning scikit-learn jupyter-notebook cnn pytorch seaborn dropout batch-normalization image-classification matplotlib resnet regularization cifar10 data-augmentation model-evaluation interpretability weight-decay train-validation-test
-
Updated
May 28, 2025 - Jupyter Notebook
Code accompanying "Learning to Forget: Continual Learning with Adaptive Weight Decay"
-
Updated
May 1, 2026 - Python
Cheap online diagnostics for grokking transformers: weight-decay regimes, attention-head order parameters, data, and Lean 4 verification
machine-learning transformers formal-verification mamba grokking state-space-models weight-decay phase-transitions lean4 mechanistic-interpretability
-
Updated
Aug 12, 2026 - Python
We proposed an order parameter for grokking. Then we killed it. A pre-registered falsification, with primary data, code, and a documented retraction.
machine-learning deep-learning reproducible-research numpy open-science modular-arithmetic research-data interpretability generalization grokking weight-decay negative-results order-parameter falsification mechanistic-interpretability
-
Updated
Jul 15, 2026 - Python
Folder contains implementation of Multi layer feed forward networks, Autoencoders, Sparse Autoencoders and many..
neural-network mnist autoencoder feedforward-neural-network regularization momentum backpropagation sparse-autoencoders sparse-autoencoder autoencoder-mnist weight-decay
-
Updated
May 17, 2019 - HTML
Super-Convergence on CIFAR10
sophia cifar10 lion second-order-optimization adamw super-convergence weight-decay sharpness-aware-minimization madgrad large-batch-optimization lion-optimizer
-
Updated
Jun 17, 2024 - Python
Implementation of some new techniques from fastai and other papers which works with keras models
-
Updated
Dec 26, 2018 - Python
Machine Learning university project
-
Updated
Apr 1, 2019 - Python
Code and figures for To Grok Grokking: Provable Grokking in Ridge Regression (ICML 2026), with ridge-regression, random-feature, NTK-style, and fully trained ReLU experiments.
pytorch gradient-descent ridge-regression machine-learning-theory generalization grokking random-features weight-decay neural-tangent-kernel overparameterization relu-networks icml-2026
-
Updated
Sep 6, 2026 - Python
Add this topic to your repo
To associate your repository with the weight-decay topic, visit your repo's landing page and select "manage topics."