AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Revisit Fuzzy Neural Network: Demystifying Batch Normalization and ReLU with Generalized Hamming Network

Lixin Fan

NeurIPS 2017 · Architectures and efficiency › RNNs and normalization · approach: Batch normalization improvements · 15 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add analytical bias rectification based on generalized Hamming distanceGHN · MNIST classificationtest accuracy: 99.0%Improved
Add double-thresholding with optimized threshold ratio (ablation)GHN · CIFAR10accuracy: 84.63% -> 89.26%Improved
Add double-thresholding with optimized threshold ratio (ablation)GHN · CIFAR100accuracy: 51.71% -> 60.05%Improved
Disable double-thresholding (ablation)GHN · MNISTtest accuracy: 98.97%, 98.86%, 98.96%, 98.69%No clear effect
Use an optimized threshold ratio instead of setting r to 0 (ablation)GHN · CIFAR10 and CIFAR100iterations to reach the same test accuracy: 1 to 2 orders of magnitudesImproved
Use analytical bias rectification rather than the baseline VAE setupMNIST VAE · MNIST reconstructionreconstruction error: about 28% reductionImproved
Use analytical bias rectificationSentence CNN · Movie Review sentence classificationevaluation accuracy: both networks yielded the final evaluation accuracy around 74%No clear effect
Use analytical bias rectificationSentence CNN · Movie Review sentence classificationlearning speedImproved
Approximate the input-dependent bias with batch normalizationneural network · MNIST first convolution layercorrelation of neuron outputs with GHD outputs: Pearson correlation coefficient=0.97Improved
Use a large learning rate of 0.1 rather than 0.01GHN · MNISTlearning speedImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy