ML Research Wiki / Benchmarks / Image Classification / MNIST

MNIST

Image Classification Benchmark

Performance Over Time

📊 Showing 74 results | 📏 Metric: Percentage error

Top Performing Models

Rank	Model	Paper	Percentage error	Date	Code
1	Branching/Merging CNN + Homogeneous Vector Capsules	No Routing Needed Between Capsules	99.87	2020-01-24	📦 AdamByerly/BMCNNwHFCs
2	EnsNet (Ensemble learning in CNN augmented with fully connected subnetworks)	Ensemble learning in CNN augmented with fully connected subnetworks	99.84	2020-03-19	📦 eaguaida/TF_EnsNet- 📦 dslisleedh/EnsNet-tensorflow2
3	Efficient-CapsNet	Efficient-CapsNet: Capsule Network with Self-Attention Routing	99.84	2021-01-29	📦 EscVM/Efficient-CapsNet 📦 kaparoo/Efficient-CapsNet
4	SOPCNN (Only a single Model)	Stochastic Optimization of Plain Convolutional Neural Networks with Simple methods	99.83	2020-01-24	📦 junaidaliop/MNIST-SOPCNN
5	RMDL (30 RDLs)	RMDL: Random Multimodel Deep Learning for Classification	99.82	2018-05-03	📦 kk7nc/RMDL
6	R-ExplaiNet-22 (single model)	Learning local discrete features in explainable-by-design convolutional neural networks	99.80	2024-10-31	📦 pikaplan/LearnExplaiNet
7	µ2Net (ViT-L/16)	An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems	99.75	2022-05-25	📦 google-research/google-research
8	VGG-5 (Spinal FC)	SpinalNet: Deep Neural Network with Gradual Input	99.72	2020-07-07	📦 dipuk0506/SpinalNet 📦 dipuk0506/uq 📦 Mechachleopteryx/SpinalNet
9	TextCaps	TextCaps : Handwritten Character Recognition with Very Small Datasets	99.71	2019-04-17	📦 vinojjayasundara/textcaps 📦 milanzongor/unihack_2020 📦 kubantjan/fast-form
10	ExquisiteNetV2	A Novel lightweight Convolutional Neural Network, ExquisiteNetV2	99.71	2021-05-19	📦 shyhyawJou/ExquisiteNetV2

All Papers (74)

No Routing Needed Between Capsules

2020

Branching/Merging CNN + Homogeneous Vector Capsules

AdamByerly/BMCNNwHFCs

Ensemble learning in CNN augmented with fully connected subnetworks

2020

EnsNet (Ensemble learning in CNN augmented with fully connected subnetworks)

eaguaida/TF_EnsNet- dslisleedh/EnsNet-tensorflow2

Efficient-CapsNet: Capsule Network with Self-Attention Routing

2021

Efficient-CapsNet

EscVM/Efficient-CapsNet kaparoo/Efficient-CapsNet

Stochastic Optimization of Plain Convolutional Neural Networks with Simple methods

2020

SOPCNN (Only a single Model)

junaidaliop/MNIST-SOPCNN

RMDL: Random Multimodel Deep Learning for Classification

2018

RMDL (30 RDLs)

kk7nc/RMDL

Learning local discrete features in explainable-by-design convolutional neural networks

2024

R-ExplaiNet-22 (single model)

pikaplan/LearnExplaiNet

An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems

2022

µ2Net (ViT-L/16)

google-research/google-research

SpinalNet: Deep Neural Network with Gradual Input

2020

VGG-5 (Spinal FC)

dipuk0506/SpinalNet dipuk0506/uq Mechachleopteryx/SpinalNet

TextCaps : Handwritten Character Recognition with Very Small Datasets

2019

TextCaps

vinojjayasundara/textcaps milanzongor/unihack_2020 kubantjan/fast-form

A Novel lightweight Convolutional Neural Network, ExquisiteNetV2

2021

ExquisiteNetV2

shyhyawJou/ExquisiteNetV2

XnODR and XnIDR: Two Accurate and Fast Fully Connected Layers For Convolutional Neural Networks

2021

MobileNet_XnODR

jiansfoggy/CODE-SHOW

CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters

2022

ResNet-9

paulgavrikov/cnn-filter-db

On Second Order Behaviour in Augmented Neural ODEs

2020

Second Order Neural Ordinary Differential Equation

a-norcliffe/sonode

Augmented Neural ODEs

2019

Augmented Neural Ordinary Differential Equation

EmilienDupont/augmented-neural-odes mitmath/18S096SciML

LR-Net: A Block-based Convolutional Neural Network for Low-Resolution Image Classification

2022

LR-Net

AshkanGanj/Block-Based-ImageClassification-Architecture

Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision

2022

SEER (RegNet10B)

facebookresearch/vissl

The Convolutional Tsetlin Machine

2019

Convolutional Tsetlin Machine

cair/TsetlinMachine cair/pyTsetlinMachine

Learning in Wilson-Cowan model for metapopulation

2024

CNN+ Wilson-Cowan model RNN

raffaelemarino/learning_in_wilsoncowan

Robust and accelerated single-spike spiking neural network training with applicability to challenging temporal tasks

2022

FastSNN (CNN)

webstorms/fastsnn

rKAN: Rational Kolmogorov-Arnold Networks

2024

rKAN

alirezaafzalaghaei/rkan

Robust Training in High Dimensions via Block Coordinate Geometric Median Descent

2021

CNN-5 Layer

anishacharya/BGMD anishacharya/Optimization-Mavericks

fKAN: Fractional Kolmogorov-Arnold Networks with trainable Jacobi basis functions

2024

fKAN

alirezaafzalaghaei/fKAN

Spike time displacement based error backpropagation in convolutional spiking neural networks

2021

StiDi-BP in R-CSNN

Parametric Matrix Models

2024

Convolutional PMM (Parametric Matrix Model)

Convolutional Sequence to Sequence Learning

2017

CNN Model by Som

facebookresearch/fairseq facebookresearch/ParlAI

The Weighted Tsetlin Machine: Compressed Representations with Weighted Clauses

2019

Weighted Tsetlin Machine

cair/pyTsetlinMachine adrianphoulady/weighted-tsetlin-machine-cpp

Tensorizing Neural Networks

2015

Perceptron with a tensor train layer

timgaripov/TensorNet-TF Bihaqo/TensorNet

Augmented Neural ODEs

2019

ANODE

EmilienDupont/augmented-neural-odes mitmath/18S096SciML

The Tsetlin Machine - A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic

2018

Tsetlin Machine

cair/TsetlinMachine cair/pyTsetlinMachine

Learning in Wilson-Cowan model for metapopulation

2024

Wilson-Cowan model RNN

raffaelemarino/learning_in_wilsoncowan

Exploring Effects of Hyperdimensional Vectors for Tsetlin Machines

2024

Hypervector Tsetlin Machine

A Single Graph Convolution Is All You Need: Efficient Grayscale Image Classification

2024

GECCO

geccoproject/gecco

Robust and accelerated single-spike spiking neural network training with applicability to challenging temporal tasks

2022

FastSNN (MLP)

webstorms/fastsnn

Parametric Matrix Models

2024

PMM (Parametric Matrix Model)

The Backpropagation Algorithm Implemented on Spiking Neuromorphic Hardware

2021

Binarized MLP with on-chip spiking backpropagation (on Loihi)

lanl/spikingBackprop

Improving k-Means Clustering Performance with Disentangled Internal Representations

2020

SNNL-5

afagarap/pt-snnl

ProjectionNet: Learning Efficient On-Device Deep Networks Using Neural Projections

2017

ProjectionNet

Personalized Federated Learning with Hidden Information on Personalized Prior

2022

pFedBreD_ns_mg

Performance of Gaussian Mixture Model Classifiers on Embedded Feature Spaces

2024

DGMMC-S

cvmlmu/dgmmc

Stacked What-Where Auto-encoders

2015

Zhao et al. (2015) (auto-encoder)

isaacgerg/keras_odds_and_ends zhangqinghao0811/unpool

On the Ideal Number of Groups for Isometric Gradient Propagation

2023

MLP (ideal number of groups)

Convolutional Clustering for Unsupervised Learning

2015

Convolutional Clustering

Sparse Networks from Scratch: Faster Training without Losing Performance

2019

LeNet 300-100 (Sparse Momentum)

TimDettmers/sparse_learning google-research/rigl

BinaryConnect: Training Deep Neural Networks with binary weights during propagations

2015

BinaryConnect

tensorpack/tensorpack MatthieuCourbariaux/BinaryConnect

Sparse Activity and Sparse Connectivity in Supervised Learning

2016

Sparse Activity and Sparse Connectivity in Supervised Learning

Explaining and Harnessing Adversarial Examples

2014

Explaining and Harnessing Adversarial Examples

cleverhans-lab/cleverhans openai/cleverhans

Deep Fried Convnets

2014

Deep Fried Convnets

v0lta/wavelet-network-compression

PCANet: A Simple Deep Learning Baseline for Image Classification?

2014

PCANet

Ldpe2G/PCANet lucasleesw/PCANet_python_ver

DiffPrune: Neural Network Pruning with Deterministic Approximate Binary Gates and $L_0$ Regularization

2020

DiffPrune (LeNet5)

YanivShu/diffprune_public

Improved Training Speed, Accuracy, and Data Utilization Through Loss Function Optimization

2019

Simple CNN with BaikalCMA loss

sgonzalez/SwiftGenetics sgonzalez/SwiftCMA

NeuPDE: Neural Network Based Ordinary and Partial Differential Equations for Modeling Time-Dependent Data

2019

NeuPDE

Maxout Networks

2013

Maxout Networks

MaximeVandegar/Papers-in-100-Lines-of-Code philipperemy/tensorflow-maxout

Network In Network

2013

NiN

MaximeVandegar/Papers-in-100-Lines-of-Code nagadomi/kaggle-cifar10-torch7

ReNet: A Recurrent Neural Network Based Alternative to Convolutional Networks

2015

ReNet

fvisin/reseg brian-moser/dartsrenet

Deep Convolutional Neural Networks as Generic Feature Extractors

2017

DCNN+GFE

Training Very Deep Networks

2015

VDN

LiyuanLucasLiu/LM-LSTM-CRF yoonkim/lstm-char-cnn flukeskywalker/highway-networks

Deeply-Supervised Nets

2014

DSN

ellisdg/3DUnetCNN

Convolutional Kernel Networks

2014

CKN

Unsupervised Feature Learning with C-SVDDNet

2014

C-SVDDNet

Hybrid Orthogonal Projection and Estimation (HOPE): A New Framework to Probe and Learn Neural Networks

2015

HOPE

Enhanced Image Classification With a Fast-Learning Shallow Convolutional Neural Network

2015

FLSCNN

On the Importance of Normalisation Layers in Deep Learning with Piecewise Linear Activation Units

2015

MIM

All you need is a good init

2015

Fitnet-LSUV-SVM

ducha-aiki/LSUVinit ducha-aiki/LSUV-keras

EXACT: How to Train Your Accuracy

2022

EXACT (M3-CNN)

tinkoff-ai/exact ivan-chai/exact

Fractional Max-Pooling

2014

Fractional MP

facebookresearch/SparseConvNet laplacetw/vgg-like-cifar10

Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree

2015

Tree+Max-Avg pooling

cypw/DPNs BeanGreen247/Python-AI-Arts

Competitive Multi-scale Convolution

2015

CMsC

WaveMix: Resource-efficient Token Mixing for Images

2022

WaveMix-128/7

pranavphoenix/WaveMix

Training Neural Networks with Local Error Signals

2019

VGG8B + LocalLearning + CO

anokland/local-loss ai-tech-research-lab/nitro-d

Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures

2016

SimpleNetv1

Coderx7/SimpleNet Coderx7/SimpleNet_Pytorch

Dynamic Routing Between Capsules

2017

CapsNet

labmlai/annotated_deep_learning_paper_implementations naturomics/CapsNet-Tensorflow

Batch-normalized Maxout Network in Network

2015

BNM NiN

JohnBensen1000/machine_learning

Multi-column Deep Neural Networks for Image Classification

2012

MCDNN

hughperkins/DeepCL

APAC: Augmented PAttern Classification with Neural Networks

2015

APAC

Model	Paper	Percentage error	Date
Branching/Merging CNN + Homogeneous Vector Capsules	No Routing Needed Between Capsules	99.87	2020-01-24
EnsNet (Ensemble learning in CNN augmented with fully connected subnetworks)	Ensemble learning in CNN augmented with fully con…	99.84	2020-03-19
Efficient-CapsNet	Efficient-CapsNet: Capsule Network with Self-Atte…	99.84	2021-01-29
SOPCNN (Only a single Model)	Stochastic Optimization of Plain Convolutional Ne…	99.83	2020-01-24
RMDL (30 RDLs)	RMDL: Random Multimodel Deep Learning for Classif…	99.82	2018-05-03
R-ExplaiNet-22 (single model)	Learning local discrete features in explainable-b…	99.80	2024-10-31
µ2Net (ViT-L/16)	An Evolutionary Approach to Dynamic Introduction …	99.75	2022-05-25
VGG-5 (Spinal FC)	SpinalNet: Deep Neural Network with Gradual Input	99.72	2020-07-07
TextCaps	TextCaps : Handwritten Character Recognition with…	99.71	2019-04-17
ExquisiteNetV2	A Novel lightweight Convolutional Neural Network,…	99.71	2021-05-19
MobileNet_XnODR	XnODR and XnIDR: Two Accurate and Fast Fully Conn…	99.68	2021-11-21
ResNet-9	CNN Filter DB: An Empirical Investigation of Trai…	99.68	2022-03-29
Second Order Neural Ordinary Differential Equation	On Second Order Behaviour in Augmented Neural ODEs	99.63	2020-06-12
Augmented Neural Ordinary Differential Equation	Augmented Neural ODEs	99.63	2019-04-02
LR-Net	LR-Net: A Block-based Convolutional Neural Networ…	99.47	2022-07-19
SEER (RegNet10B)	Vision Models Are More Robust And Fair When Pretr…	99.42	2022-02-16
Convolutional Tsetlin Machine	The Convolutional Tsetlin Machine	99.40	2019-05-23
CNN+ Wilson-Cowan model RNN	Learning in Wilson-Cowan model for metapopulation	99.31	2024-06-24
FastSNN (CNN)	Robust and accelerated single-spike spiking neura…	99.30	2022-05-30
rKAN	rKAN: Rational Kolmogorov-Arnold Networks	99.29	2024-06-20
CNN-5 Layer	Robust Training in High Dimensions via Block Coor…	99.27	2021-06-16
fKAN	fKAN: Fractional Kolmogorov-Arnold Networks with …	99.23	2024-06-11
StiDi-BP in R-CSNN	Spike time displacement based error backpropagati…	99.20	2021-08-31
Convolutional PMM (Parametric Matrix Model)	Parametric Matrix Models	98.99	2024-01-22
CNN Model by Som	Convolutional Sequence to Sequence Learning	98.59	2017-05-08
Weighted Tsetlin Machine	The Weighted Tsetlin Machine: Compressed Represen…	98.50	2019-11-28
Perceptron with a tensor train layer	Tensorizing Neural Networks	98.20	2015-09-22
ANODE	Augmented Neural ODEs	98.20	2019-04-02
Tsetlin Machine	The Tsetlin Machine - A Game Theoretic Bandit Dri…	98.20	2018-04-04
Wilson-Cowan model RNN	Learning in Wilson-Cowan model for metapopulation	98.13	2024-06-24
Hypervector Tsetlin Machine	Exploring Effects of Hyperdimensional Vectors for…	98.13	2024-06-04
GECCO	A Single Graph Convolution Is All You Need: Effic…	98.04	2024-02-01
FastSNN (MLP)	Robust and accelerated single-spike spiking neura…	97.91	2022-05-30
PMM (Parametric Matrix Model)	Parametric Matrix Models	97.38	2024-01-22
Binarized MLP with on-chip spiking backpropagation (on Loihi)	The Backpropagation Algorithm Implemented on Spik…	96.20	2021-06-13
SNNL-5	Improving k-Means Clustering Performance with Dis…	95.50	2020-06-05
ProjectionNet	ProjectionNet: Learning Efficient On-Device Deep …	95.00	2017-08-02
pFedBreD_ns_mg	Personalized Federated Learning with Hidden Infor…	92.47	2022-11-19
DGMMC-S	Performance of Gaussian Mixture Model Classifiers…	70.00	2024-10-17
Zhao et al. (2015) (auto-encoder)	Stacked What-Where Auto-encoders	4.76	2015-06-08
MLP (ideal number of groups)	On the Ideal Number of Groups for Isometric Gradi…	1.67	2023-02-07
Convolutional Clustering	Convolutional Clustering for Unsupervised Learning	1.40	2015-11-19
LeNet 300-100 (Sparse Momentum)	Sparse Networks from Scratch: Faster Training wit…	1.26	2019-07-10
BinaryConnect	BinaryConnect: Training Deep Neural Networks with…	1.00	2015-11-02
Sparse Activity and Sparse Connectivity in Supervised Learning	Sparse Activity and Sparse Connectivity in Superv…	0.80	2016-03-28
Explaining and Harnessing Adversarial Examples	Explaining and Harnessing Adversarial Examples	0.80	2014-12-20
Deep Fried Convnets	Deep Fried Convnets	0.70	2014-12-22
PCANet	PCANet: A Simple Deep Learning Baseline for Image…	0.60	2014-04-14
DiffPrune (LeNet5)	DiffPrune: Neural Network Pruning with Determinis…	0.60	2020-12-07
Simple CNN with BaikalCMA loss	Improved Training Speed, Accuracy, and Data Utili…	0.53	2019-05-27
NeuPDE	NeuPDE: Neural Network Based Ordinary and Partial…	0.51	2019-08-08
Maxout Networks	Maxout Networks	0.50	2013-02-18
NiN	Network In Network	0.50	2013-12-16
ReNet	ReNet: A Recurrent Neural Network Based Alternati…	0.50	2015-05-03
DCNN+GFE	Deep Convolutional Neural Networks as Generic Fea…	0.50	2017-10-06
VDN	Training Very Deep Networks	0.50	2015-07-22
DSN	Deeply-Supervised Nets	0.40	2014-09-18
CKN	Convolutional Kernel Networks	0.40	2014-06-12
C-SVDDNet	Unsupervised Feature Learning with C-SVDDNet	0.40	2014-12-23
HOPE	Hybrid Orthogonal Projection and Estimation (HOPE…	0.40	2015-02-03
FLSCNN	Enhanced Image Classification With a Fast-Learnin…	0.40	2015-03-16
MIM	On the Importance of Normalisation Layers in Deep…	0.40	2015-08-03
Fitnet-LSUV-SVM	All you need is a good init	0.40	2015-11-19
EXACT (M3-CNN)	EXACT: How to Train Your Accuracy	0.33	2022-05-19
Fractional MP	Fractional Max-Pooling	0.30	2014-12-18
Tree+Max-Avg pooling	Generalizing Pooling Functions in Convolutional N…	0.30	2015-09-30
CMsC	Competitive Multi-scale Convolution	0.30	2015-11-18
WaveMix-128/7	WaveMix: Resource-efficient Token Mixing for Imag…	0.29	2022-03-07
VGG8B + LocalLearning + CO	Training Neural Networks with Local Error Signals	0.26	2019-01-20
SimpleNetv1	Lets keep it simple, Using simple architectures t…	0.25	2016-08-22
CapsNet	Dynamic Routing Between Capsules	0.25	2017-10-26
BNM NiN	Batch-normalized Maxout Network in Network	0.24	2015-11-09
MCDNN	Multi-column Deep Neural Networks for Image Class…	0.23	2012-02-13
APAC	APAC: Augmented PAttern Classification with Neura…	0.23	2015-05-13

MNIST

Performance Over Time

Edit Benchmark Results

Edit Result

Top Performing Models

All Papers (74)