Image Classification on CIFAR-100

Metric: Percentage correct (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Percentage correct▼	Extra Data	Paper	Date↕	Code
1	EffNet-L2 (SAM)	96.08	Yes	Sharpness-Aware Minimization for Efficiently Imp...	2020-10-03	Code
2	Swin-L + ML-Decoder	95.1	Yes	ML-Decoder: Scalable and Versatile Classificatio...	2021-11-25	Code
3	µ2Net (ViT-L/16)	94.95	Yes	An Evolutionary Approach to Dynamic Introduction...	2022-05-25	Code
4	ViT-B-16 (ImageNet-21K-P pretrain)	94.2	Yes	ImageNet-21K Pretraining for the Masses	2021-04-22	Code
5	CvT-W24	94.09	Yes	CvT: Introducing Convolutions to Vision Transfor...	2021-03-29	Code
6	ViT-B/16 (PUGD)	93.95	Yes	Perturbated Gradients Updating within Unit Space...	2021-10-01	Code
7	Heinsen Routing + BEiT-large 16 224	93.8	Yes	An Algorithm for Routing Vectors in Sequences	2022-11-20	Code
8	BiT-L (ResNet)	93.51	Yes	Big Transfer (BiT): General Visual Representatio...	2019-12-24	Code
9	VIT-L/16 (Spinal FC, Background)	93.31	No	Reduction of Class Activation Uncertainty with B...	2023-05-05	Code
10	CaiT-M-36 U 224	93.1	Yes	Going deeper with Image Transformers	2021-03-31	Code
11	ViT-L (attn fine-tune)	93	Yes	Three things everyone should know about Vision T...	2022-03-18	Code
12	TResNet-L-V2	92.6	Yes	TResNet: High Performance GPU-Dedicated Architec...	2020-03-30	Code
13	EfficientNetV2-L	92.3	Yes	EfficientNetV2: Smaller Models and Faster Training	2021-04-01	Code
14	EfficientNetV2-M	92.2	Yes	EfficientNetV2: Smaller Models and Faster Training	2021-04-01	Code
15	BiT-M (ResNet)	92.17	Yes	Big Transfer (BiT): General Visual Representatio...	2019-12-24	Code
16	CeiT-S	91.8	Yes	Incorporating Convolution Designs into Visual Tr...	2021-03-22	Code
17	CeiT-S (384 finetune resolution)	91.8	Yes	Incorporating Convolution Designs into Visual Tr...	2021-03-22	Code
18	EfficientNet-B7	91.7	Yes	EfficientNet: Rethinking Model Scaling for Convo...	2019-05-28	Code
19	EfficientNetV2-S	91.5	Yes	EfficientNetV2: Smaller Models and Faster Training	2021-04-01	Code
20	GPIPE	91.3	Yes	GPipe: Efficient Training of Giant Neural Networ...	2018-11-16	Code
21	TNT-B	91.1	Yes	Transformer in Transformer	2021-02-27	Code
22	DeiT-B	90.8	Yes	Training data-efficient image transformers & dis...	2020-12-23	Code
23	GFNet-H-B	90.3	Yes	Global Filter Networks for Image Classification	2021-07-01	Code
24	E2E-3M	90.27	Yes	Rethinking Recurrent Neural Networks and Other I...	2020-07-30	Code
25	Bamboo (ViT-B/16)	90.2	Yes	Bamboo: Building Mega-Scale Vision Dataset Conti...	2022-03-15	Code
26	PyramidNet-272 (ASAM)	89.9	No	ASAM: Adaptive Sharpness-Aware Minimization for ...	2021-02-23	Code
27	PyramidNet (SAM)	89.7	No	Sharpness-Aware Minimization for Efficiently Imp...	2020-10-03	Code
28	DVT (T2T-ViT-24)	89.63	Yes	Not All Images are Worth 16x16 Words: Dynamic Tr...	2021-05-31	Code
29	ResMLP-24	89.5	Yes	ResMLP: Feedforward networks for image classific...	2021-05-07	Code
30	PyramidNet-272, S=4	89.46	Yes	Towards Better Accuracy-efficiency Trade-offs: D...	2020-11-30	Code
31	CeiT-T	89.4	Yes	Incorporating Convolution Designs into Visual Tr...	2021-03-22	Code
32	PyramidNet+ShakeDrop	89.3	Yes	AutoAugment: Learning Augmentation Policies from...	2018-05-24	Code
33	ViT-B/16- SAM	89.1	Yes	When Vision Transformers Outperform ResNets with...	2021-06-03	Code
34	ConvMLP-M	89.1	Yes	ConvMLP: Hierarchical Convolutional MLPs for Vis...	2021-09-09	Code
35	ConvMLP-L	88.6	Yes	ConvMLP: Hierarchical Convolutional MLPs for Vis...	2021-09-09	Code
36	ResNet-152x4-AGC (ImageNet-21K)	88.54	Yes	Effect of Pre-Training Scale on Intra- and Inter...	2021-05-31	Code
37	ColorNet	88.4	Yes	ColorNet: Investigating the importance of color ...	2019-02-01	Code
38	PyramidNet+ShakeDrop (Fast AA)	88.3	Yes	Fast AutoAugment	2019-05-01	Code
39	NAT-M4	88.3	Yes	Neural Architecture Transfer	2020-05-12	Code
40	CeiT-T (384 finetune resolution)	88	Yes	Incorporating Convolution Designs into Visual Tr...	2021-03-22	Code
41	NAT-M3	87.7	Yes	Neural Architecture Transfer	2020-05-12	Code
42	ViT-S/16- SAM	87.6	Yes	When Vision Transformers Outperform ResNets with...	2021-06-03	Code
43	NAT-M2	87.5	Yes	Neural Architecture Transfer	2020-05-12	Code
44	Dynamics 1	87.48	Yes	PSO-Convolutional Neural Networks with Heterogen...	2022-05-20	Code
45	DenseNet-BC-190, S=4	87.44	Yes	Towards Better Accuracy-efficiency Trade-offs: D...	2020-11-30	Code
46	ConvMLP-S	87.4	Yes	ConvMLP: Hierarchical Convolutional MLPs for Vis...	2021-09-09	Code
47	ResMLP-12	87	Yes	ResMLP: Feedforward networks for image classific...	2021-05-07	Code
48	WRN-40-10, S=4	86.9	Yes	Towards Better Accuracy-efficiency Trade-offs: D...	2020-11-30	Code
49	ResNet50 (A1)	86.9	Yes	ResNet strikes back: An improved training proced...	2021-10-01	Code
50	WRN-28-10 * 3	86.81	Yes	MixMo: Mixing Multiple Inputs for Multiple Outpu...	2021-03-10	Code
51	PyramidNet + AA (AMP)	86.64	Yes	Regularizing Neural Networks via Adversarial Mod...	2020-10-10	Code
52	PyramidNet-200 + Shakedrop + Cutmix + PS-KD	86.41	Yes	Self-Knowledge Distillation with Progressive Ref...	2020-06-22	Code
53	Mixer-B/16- SAM	86.4	Yes	When Vision Transformers Outperform ResNets with...	2021-06-03	Code
54	ResCNet-50	86.31	No	Deep Feature Response Discriminative Calibration	2024-11-16	Code
55	PyramidNet-200 + Shakedrop + Cutmix	86.19	Yes	CutMix: Regularization Strategy to Train Strong ...	2019-05-13	Code
56	MUXNet-m	86.1	Yes	MUXConv: Information Multiplexing in Convolution...	2020-03-31	Code
57	NAT-M1	86	Yes	Neural Architecture Transfer	2020-05-12	Code
58	WRN-28-10	85.77	Yes	MixMo: Mixing Multiple Inputs for Multiple Outpu...	2021-03-10	Code
59	WRN-28-10, S=4	85.74	Yes	Towards Better Accuracy-efficiency Trade-offs: D...	2020-11-30	Code
60	WRN-28-8 (SAMix+DM)	85.59	No	-	-	-
61	WRN-28-8 +SAMix	85.5	Yes	Boosting Discriminative Visual Representation Le...	2021-11-30	Code
62	ASANas	85.42	Yes	Improving Neural Architecture Search Image Class...	2019-03-14	Code
63	WRN-28-8 (AutoMix+DM)	85.38	No	-	-	-
64	SparseSwin	85.35	Yes	SparseSwin: Swin Transformer with Sparse Transfo...	2023-09-11	Code
65	WRN-28-8 (PuzzleMix+DM)	85.25	No	-	-	-
66	ResNet-50-SAM	85.2	Yes	When Vision Transformers Outperform ResNets with...	2021-06-03	Code
67	WRN-28-8 +AutoMix	85.16	Yes	AutoMix: Unveiling the Power of Mixup for Strong...	2021-03-24	Code
68	WaveMixLite-256/7	85.09	Yes	WaveMix: A Resource-efficient Neural Network for...	2022-05-28	Code
69	MANO-tiny	85.08	Yes	Linear Attention with Global Context: A Multipol...	2025-07-03	Code
70	WRN 28-14	85	Yes	Neural networks with late-phase weights	2020-07-25	Code
71	R-Mix (WideResNet 28-10)	85	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
72	EEEA-Net-C (b=5)+ CO	84.98	Yes	EEEA-Net: An Early Exit Evolutionary Neural Arch...	2021-08-13	Code
73	RL-Mix (WideResNet 28-10)	84.9	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
74	Wide-ResNet-28-10	84.89	Yes	Automatic Data Augmentation via Invariance-Const...	2022-09-29	Code
75	SENet + ShakeEven + Cutout	84.59	Yes	Squeeze-and-Excitation Networks	2017-09-05	Code
76	ResNeXt-50(32x4d) + SAMix	84.42	Yes	Boosting Discriminative Visual Representation Le...	2021-11-30	Code
77	WRN-28-10 with reSGHMC	84.38	Yes	Non-convex Learning via Replica Exchange Stochas...	2020-08-12	Code
78	PyramidNet-272 + SWA	84.16	Yes	Averaging Weights Leads to Wider Optima and Bett...	2018-03-14	Code
79	WRN28-10	84.05	Yes	Puzzle Mix: Exploiting Saliency and Local Statis...	2020-09-15	Code
80	HCGNet-A3	84.04	Yes	Gated Convolutional Networks with Hybrid Connect...	2019-08-26	Code
81	WideResNet 28-10 + CutMix (OneCycleLR scheduler)	83.97	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
82	DenseNet-BC-190 + FMix	83.95	Yes	FMix: Enhancing Mixed Sample Data Augmentation	2020-02-27	Code
83	ORN	83.85	Yes	Oriented Response Networks	2017-01-07	Code
84	Grafit (ResNet-50)	83.7	Yes	Grafit: Learning fine-grained image representati...	2020-11-25	-
85	ResNeXt-50(32x4d) + AutoMix	83.64	Yes	AutoMix: Unveiling the Power of Mixup for Strong...	2021-03-24	Code
86	CCT-7/3x1+HTM+VTM	83.57	Yes	TokenMixup: Efficient Attention-guided Token-lev...	2022-10-14	Code
87	HCGNet-A2	83.46	Yes	Gated Convolutional Networks with Hybrid Connect...	2019-08-26	Code
88	Res2NeXt-29	83.44	Yes	Res2Net: A New Multi-scale Backbone Architecture	2019-04-02	Code
89	DenseNet-BC-190 + Mixup	83.2	Yes	mixup: Beyond Empirical Risk Minimization	2017-10-25	Code
90	SSAL-DenseNet 190-40	83.2	Yes	Contextual Classification Using Self-Supervised ...	2021-01-07	Code
91	EnAET	83.13	Yes	EnAET: A Self-Trained framework for Semi-Supervi...	2019-11-21	Code
92	WRN 28-10	83.06	Yes	Neural networks with late-phase weights	2020-07-25	Code
93	R-Mix (ResNeXt 29-4-24)	83.02	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
94	Wide ResNet+Cutout+no BN scale/offset learning	82.95	Yes	Single-bit-per-weight deep convolutional neural ...	2019-07-16	Code
95	WRN-16-8 with reSGHMC	82.95	Yes	Non-convex Learning via Replica Exchange Stochas...	2020-08-12	Code
96	DenseNet-BC	82.82	Yes	Densely Connected Convolutional Networks	2016-08-25	Code
97	ABNet-2G-R3-Combined	82.784	No	ANDHRA Bandersnatch: Training Neural Networks to...	2024-11-28	Code
98	CCT-7/3x1*	82.72	Yes	Escaping the Big Data Paradigm with Compact Tran...	2021-04-12	Code
99	EXACT (WRN-28-10)	82.68	No	EXACT: How to Train Your Accuracy	2022-05-19	Code
100	SKNet-29 (ResNeXt-29, 16×32d)	82.67	Yes	Selective Kernel Networks	2019-03-15	Code
101	DenseNet	82.62	Yes	Densely Connected Convolutional Networks	2016-08-25	Code
102	Shared WRN	82.57	Yes	Learning Implicitly Recurrent CNNs Through Param...	2019-02-26	Code
103	Transformer local-attention (NesT-B)	82.56	Yes	Nested Hierarchical Transformer: Towards Accurat...	2021-05-26	Code
104	RL-Mix (ResNeXt 29-4-24)	82.43	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
105	Mixer-S/16- SAM	82.4	Yes	When Vision Transformers Outperform ResNets with...	2021-06-03	Code
106	R-Mix (WideResNet 16-8)	82.32	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
107	ResNeXt 29-4-24 + CutMix (OneCycleLR scheduler)	82.3	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
108	WARN	82.18	Yes	Attend and Rectify: a Gated Attention Mechanism ...	2018-07-19	Code
109	RL-Mix (WideResNet 16-8)	82.16	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
110	WRN+SWA	82.15	Yes	Averaging Weights Leads to Wider Optima and Bett...	2018-03-14	Code
111	Manifold Mixup	81.96	Yes	Manifold Mixup: Better Representations by Interp...	2018-06-13	Code
112	HCGNet-A1	81.87	Yes	Gated Convolutional Networks with Hybrid Connect...	2019-08-26	Code
113	WideResNet 16-8 + CutMix (OneCycleLR scheduler)	81.79	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
114	Residual Gates + WRN	81.73	Yes	Learning Identity Mappings with Residual Gates	2016-11-04	-
115	kNN-CLIP	81.7	Yes	Revisiting a kNN-based Image Classification Syst...	2022-04-03	-
116	AA-Wide-ResNet	81.6	Yes	Attention Augmented Convolutional Networks	2019-04-22	Code
117	PDO-eConv (p8, 4.6M)	81.6	Yes	PDO-eConvs: Partial Differential Operator Based ...	2020-07-20	Code
118	SEER (RegNet10B)	81.53	Yes	Vision Models Are More Robust And Fair When Pret...	2022-02-16	Code
119	R-Mix (PreActResNet-18)	81.49	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
120	ResNet50 (FSGDM)	81.44	No	On the Performance Analysis of Momentum Method: ...	2024-11-29	Code
121	Wide-ResNet-40-2	81.19	Yes	Automatic Data Augmentation via Invariance-Const...	2022-09-29	Code
122	Wide ResNet	81.15	Yes	Wide Residual Networks	2016-05-23	Code
123	CoPaNet-R-164	81.1	Yes	Deep Competitive Pathway Networks	2017-09-29	Code
124	ABNet-2G-R3	80.83	No	ANDHRA Bandersnatch: Training Neural Networks to...	2024-11-28	Code
125	RL-Mix (PreActResNet-18)	80.75	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
126	PreActResNet-18 + CutMix (OneCycleLR scheduler)	80.6	Yes	Expeditious Saliency-guided Mix-up through Rando...	2022-12-09	Code
127	GAC-SNN	80.45	No	Gated Attention Coding for Training High-perform...	2023-08-12	Code
128	ABNet-2G-R2	80.354	No	ANDHRA Bandersnatch: Training Neural Networks to...	2024-11-28	Code
129	SimpleNetv2	80.29	Yes	Towards Principled Design of Deep Convolutional ...	2018-02-17	Code
130	UPANets	80.29	Yes	UPANets: Learning from the Universal Pixel Atten...	2021-03-15	Code
131	PreActResNet-18 + SageMix	80.16	Yes	SageMix: Saliency-Guided Mixup for Point Clouds	2022-10-13	Code
132	ResNet56 with reSGHMC	80.14	Yes	Non-convex Learning via Replica Exchange Stochas...	2020-08-12	Code
133	PDO-eConv (p8, 2.62M)	79.99	Yes	PDO-eConvs: Partial Differential Operator Based ...	2020-07-20	Code
134	VGG11B(3x) + LocalLearning	79.9	Yes	Training Neural Networks with Local Error Signals	2019-01-20	Code
135	NNCLR	79	Yes	With a Little Help from My Friends: Nearest-Neig...	2021-04-29	Code
136	ABNet-2G-R1	78.792	No	ANDHRA Bandersnatch: Training Neural Networks to...	2024-11-28	Code
137	PreActResNet18 (AMP)	78.49	Yes	Regularizing Neural Networks via Adversarial Mod...	2020-10-10	Code
138	SimpleNetv1	78.37	Yes	Lets keep it simple, Using simple architectures ...	2016-08-22	Code
139	ViT (lightweight, MAE pre-trained)	78.27	No	Pre-training of Lightweight Vision Transformers ...	2024-02-06	-
140	PDC	77.9	Yes	Augmenting Deep Classifiers with Polynomial Neur...	2021-04-16	Code
141	MobileNetV3-large x1.0 (BSConv-U)	77.7	Yes	Rethinking Depthwise Separable Convolutions: How...	2020-03-30	Code
142	CCT-6/3x1	77.31	Yes	Escaping the Big Data Paradigm with Compact Tran...	2021-04-12	Code
143	ResNet-1001	77.3	Yes	Identity Mappings in Deep Residual Networks	2016-03-16	Code
144	Evolution	77	Yes	Large-Scale Evolution of Image Classifiers	2017-03-03	Code
145	DIANet	76.98	Yes	DIANet: Dense-and-Implicit Attention Network	2019-05-25	Code
146	LP-BNN (ours) + cutout	76.85	Yes	Encoding the latent posterior of Bayesian Neural...	2020-12-04	Code
147	ResNet-18+MM+FRL	76.64	Yes	Learning Class Unique Features in Fine-Grained V...	2020-11-22	-
148	ResNet32 with reSGHMC	76.55	Yes	Non-convex Learning via Replica Exchange Stochas...	2020-08-12	Code
149	MomentumNet	76.38	Yes	Momentum Residual Neural Networks	2021-02-15	Code
150	SSCNN	75.7	Yes	Spatially-sparse convolutional neural networks	2014-09-22	Code
151	Exponential Linear Units	75.7	Yes	Fast and Accurate Deep Network Learning by Expon...	2015-11-23	Code
152	ResNet-9	75.59	Yes	CNN Filter DB: An Empirical Investigation of Tra...	2022-03-29	Code
153	Stochastic Depth	75.42	Yes	Deep Networks with Stochastic Depth	2016-03-30	Code
154	ResNet v2-110 (Mish activation)	74.41	Yes	Mish: A Self Regularized Non-Monotonic Activatio...	2019-08-23	Code
155	Dspike (ResNet-18)	74.24	No	-	-	-
156	ResNet20 with reSGHMC	74.14	Yes	Non-convex Learning via Replica Exchange Stochas...	2020-08-12	Code
157	MixMatch	74.1	Yes	MixMatch: A Holistic Approach to Semi-Supervised...	2019-05-06	Code
158	Beta-Rank	74.01	No	Beta-Rank: A Robust Convolutional Filter Pruning...	2023-04-15	Code
159	PreResNet-110	73.98	No	How to Use Dropout Correctly on Residual Network...	2023-02-13	Code
160	ABNet-2G-R0	73.93	No	ANDHRA Bandersnatch: Training Neural Networks to...	2024-11-28	Code
161	Fractional MP	73.6	Yes	Fractional Max-Pooling	2014-12-18	Code
162	ResNet+ELU	73.5	No	Deep Residual Networks with Exponential Linear U...	2016-04-14	Code
163	PDO-eConv (p6m,0.37M)	73	Yes	PDO-eConvs: Partial Differential Operator Based ...	2020-07-20	Code
164	SOPCNN	72.96	Yes	Stochastic Optimization of Plain Convolutional N...	2020-01-24	Code
165	PDO-eConv (p6,0.36M)	72.87	Yes	PDO-eConvs: Partial Differential Operator Based ...	2020-07-20	Code
166	Tuned CNN	72.6	Yes	Scalable Bayesian Optimization Using Deep Neural...	2015-02-19	Code
167	ResNet-110 (SAP)	72.537	No	Stochastic Subsampling With Average Pooling	2024-09-25	-
168	CMsC	72.4	Yes	Competitive Multi-scale Convolution	2015-11-18	-
169	Fitnet4-LSUV	72.3	Yes	All you need is a good init	2015-11-19	Code
170	GAN+ResNet	71.52	No	-	-	Code
171	kMobileNet V3 Large 16ch	71.36	Yes	-	-	Code
172	BNM NiN	71.1	Yes	Batch-normalized Maxout Network in Network	2015-11-09	Code
173	OTTT	71.05	No	Online Training Through Time for Spiking Neural ...	2022-10-09	Code
174	MIM	70.8	Yes	On the Importance of Normalisation Layers in Dee...	2015-08-03	-
175	WaveMix-Lite-256/7	70.2	No	WaveMix: A Resource-efficient Neural Network for...	2022-05-28	Code
176	IM-Loss (VGG-16)	70.18	No	-	-	-
177	NiN+APL	69.2	Yes	Learning Activation Functions to Improve Deep Ne...	2014-12-21	Code
178	SWWAE	69.1	Yes	Stacked What-Where Auto-encoders	2015-06-08	Code
179	NiN+Superclass+CDJ	69	Yes	Deep Convolutional Decision Jungle for Image Cla...	2017-06-06	-
180	Spectral Representations for Convolutional Neural Networks	68.4	No	Spectral Representations for Convolutional Neura...	2015-06-11	-
181	ReActNet-18	68.34	No	"BNN - BN = ?": Training Binary Neural Networks ...	2021-04-16	Code
182	VDN	67.8	No	Training Very Deep Networks	2015-07-22	Code
183	DCNN+GFE	67.7	No	Deep Convolutional Neural Networks as Generic Fe...	2017-10-06	-
184	Tree+Max-Avg pooling	67.6	No	Generalizing Pooling Functions in Convolutional ...	2015-09-30	Code
185	HD-CNN	67.4	No	HD-CNN: Hierarchical Deep Convolutional Neural N...	2014-10-03	Code
186	Universum Prescription	67.2	No	Universum Prescription: Regularization using Unl...	2015-11-11	-
187	ResNet50 Without Transfer Learning	67.06	No	-	-	Code
188	AlexNet (KP)	66.78	No	-	-	-
189	ACN	66.3	No	Striving for Simplicity: The All Convolutional Net	2014-12-21	Code
190	DLME (ResNet-18, linear)	66.1	No	DLME: Deep Local-flatness Manifold Embedding	2022-07-07	Code
191	ResNet-18 (modified)	66	No	FatNet: High Resolution Kernels for Classificati...	2022-10-30	Code
192	DSN	65.4	No	Deeply-Supervised Nets	2014-09-18	Code
193	NiN	64.3	No	Network In Network	2013-12-16	Code
194	Tree Priors	63.2	No	-	-	-
195	DNN+Probabilistic Maxout	61.9	No	Improving Deep Neural Networks with Probabilisti...	2013-12-20	-
196	Maxout Network (k=2)	61.43	No	Maxout Networks	2013-02-18	Code
197	ResNet20+UnsharpMaskLayer	60.36	No	-	-	Code
198	Convolutional Linear Transformer for Vision (CLTV)	60.11	No	Convolutional Xformers for Vision	2022-01-25	Code
199	FatNet of ResNet-18	60	No	FatNet: High Resolution Kernels for Classificati...	2022-10-30	Code
200	Optical Simulation of FatNet	60	No	FatNet: High Resolution Kernels for Classificati...	2022-10-30	Code
201	RReLU	59.8	No	Empirical Evaluation of Rectified Activations in...	2015-05-05	Code
202	Stochastic Pooling	57.5	No	Stochastic Pooling for Regularization of Deep Co...	2013-01-16	Code
203	Sign-symmetry	48.75	No	How Important is Weight Symmetry in Backpropagat...	2015-10-17	Code
204	AlexNet (DFA)	48.03	No	-	-	-
205	CNN39	42.64	No	Sharpness-Aware Minimization for Efficiently Imp...	2020-10-03	Code
206	CNN36	36.07	No	Sharpness-Aware Minimization for Efficiently Imp...	2020-10-03	Code
207	CNN37	35.05	No	Sharpness-aware Quantization for Deep Neural Net...	2021-11-24	Code
208	AlexNet (FA)	19.49	No	-	-	-

#1EffNet-L2 (SAM)SOTA
96.08
Percentage correct· Extra Data· 2020-10-03
Sharpness-Aware Minimization for Efficiently Improving Generalization Code
#2Swin-L + ML-Decoder
95.1
Percentage correct· Extra Data· 2021-11-25
ML-Decoder: Scalable and Versatile Classification Head Code
#3µ2Net (ViT-L/16)
94.95
Percentage correct· Extra Data· 2022-05-25
An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems Code
#4ViT-B-16 (ImageNet-21K-P pretrain)
94.2
Percentage correct· Extra Data· 2021-04-22
ImageNet-21K Pretraining for the Masses Code
#5CvT-W24
94.09
Percentage correct· Extra Data· 2021-03-29
CvT: Introducing Convolutions to Vision Transformers Code
#6ViT-B/16 (PUGD)
93.95
Percentage correct· Extra Data· 2021-10-01
Perturbated Gradients Updating within Unit Space for Deep Learning Code
#7Heinsen Routing + BEiT-large 16 224
93.8
Percentage correct· Extra Data· 2022-11-20
An Algorithm for Routing Vectors in Sequences Code
#8BiT-L (ResNet)SOTA
93.51
Percentage correct· Extra Data· 2019-12-24
Big Transfer (BiT): General Visual Representation Learning Code
#9VIT-L/16 (Spinal FC, Background)
93.31
Percentage correct· 2023-05-05
Reduction of Class Activation Uncertainty with Background Information Code
#10CaiT-M-36 U 224
93.1
Percentage correct· Extra Data· 2021-03-31
Going deeper with Image Transformers Code
#11ViT-L (attn fine-tune)
93
Percentage correct· Extra Data· 2022-03-18
Three things everyone should know about Vision Transformers Code
#12TResNet-L-V2
92.6
Percentage correct· Extra Data· 2020-03-30
TResNet: High Performance GPU-Dedicated Architecture Code
#13EfficientNetV2-L
92.3
Percentage correct· Extra Data· 2021-04-01
EfficientNetV2: Smaller Models and Faster Training Code
#14EfficientNetV2-M
92.2
Percentage correct· Extra Data· 2021-04-01
EfficientNetV2: Smaller Models and Faster Training Code
#15BiT-M (ResNet)
92.17
Percentage correct· Extra Data· 2019-12-24
Big Transfer (BiT): General Visual Representation Learning Code
#16CeiT-S
91.8
Percentage correct· Extra Data· 2021-03-22
Incorporating Convolution Designs into Visual Transformers Code
#17CeiT-S (384 finetune resolution)
91.8
Percentage correct· Extra Data· 2021-03-22
Incorporating Convolution Designs into Visual Transformers Code
#18EfficientNet-B7SOTA
91.7
Percentage correct· Extra Data· 2019-05-28
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks Code
#19EfficientNetV2-S
91.5
Percentage correct· Extra Data· 2021-04-01
EfficientNetV2: Smaller Models and Faster Training Code
#20GPIPESOTA
91.3
Percentage correct· Extra Data· 2018-11-16
GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism Code
#21TNT-B
91.1
Percentage correct· Extra Data· 2021-02-27
Transformer in Transformer Code
#22DeiT-B
90.8
Percentage correct· Extra Data· 2020-12-23
Training data-efficient image transformers & distillation through attention Code
#23GFNet-H-B
90.3
Percentage correct· Extra Data· 2021-07-01
Global Filter Networks for Image Classification Code
#24E2E-3M
90.27
Percentage correct· Extra Data· 2020-07-30
Rethinking Recurrent Neural Networks and Other Improvements for Image Classification Code
#25Bamboo (ViT-B/16)
90.2
Percentage correct· Extra Data· 2022-03-15
Bamboo: Building Mega-Scale Vision Dataset Continually with Human-Machine Synergy Code
#26PyramidNet-272 (ASAM)
89.9
Percentage correct· 2021-02-23
ASAM: Adaptive Sharpness-Aware Minimization for Scale-Invariant Learning of Deep Neural Networks Code
#27PyramidNet (SAM)
89.7
Percentage correct· 2020-10-03
Sharpness-Aware Minimization for Efficiently Improving Generalization Code
#28DVT (T2T-ViT-24)
89.63
Percentage correct· Extra Data· 2021-05-31
Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognition Code
#29ResMLP-24
89.5
Percentage correct· Extra Data· 2021-05-07
ResMLP: Feedforward networks for image classification with data-efficient training Code
#30PyramidNet-272, S=4
89.46
Percentage correct· Extra Data· 2020-11-30
Towards Better Accuracy-efficiency Trade-offs: Divide and Co-training Code
#31CeiT-T
89.4
Percentage correct· Extra Data· 2021-03-22
Incorporating Convolution Designs into Visual Transformers Code
#32PyramidNet+ShakeDropSOTA
89.3
Percentage correct· Extra Data· 2018-05-24
AutoAugment: Learning Augmentation Policies from Data Code
#33ViT-B/16- SAM
89.1
Percentage correct· Extra Data· 2021-06-03
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations Code
#34ConvMLP-M
89.1
Percentage correct· Extra Data· 2021-09-09
ConvMLP: Hierarchical Convolutional MLPs for Vision Code
#35ConvMLP-L
88.6
Percentage correct· Extra Data· 2021-09-09
ConvMLP: Hierarchical Convolutional MLPs for Vision Code
#36ResNet-152x4-AGC (ImageNet-21K)
88.54
Percentage correct· Extra Data· 2021-05-31
Effect of Pre-Training Scale on Intra- and Inter-Domain Full and Few-Shot Transfer Learning for Natural and Medical X-Ray Chest Images Code
#37ColorNet
88.4
Percentage correct· Extra Data· 2019-02-01
ColorNet: Investigating the importance of color spaces for image classification Code
#38PyramidNet+ShakeDrop (Fast AA)
88.3
Percentage correct· Extra Data· 2019-05-01
Fast AutoAugment Code
#39NAT-M4
88.3
Percentage correct· Extra Data· 2020-05-12
Neural Architecture Transfer Code
#40CeiT-T (384 finetune resolution)
88
Percentage correct· Extra Data· 2021-03-22
Incorporating Convolution Designs into Visual Transformers Code
#41NAT-M3
87.7
Percentage correct· Extra Data· 2020-05-12
Neural Architecture Transfer Code
#42ViT-S/16- SAM
87.6
Percentage correct· Extra Data· 2021-06-03
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations Code
#43NAT-M2
87.5
Percentage correct· Extra Data· 2020-05-12
Neural Architecture Transfer Code
#44Dynamics 1
87.48
Percentage correct· Extra Data· 2022-05-20
PSO-Convolutional Neural Networks with Heterogeneous Learning Rate Code
#45DenseNet-BC-190, S=4
87.44
Percentage correct· Extra Data· 2020-11-30
Towards Better Accuracy-efficiency Trade-offs: Divide and Co-training Code
#46ConvMLP-S
87.4
Percentage correct· Extra Data· 2021-09-09
ConvMLP: Hierarchical Convolutional MLPs for Vision Code
#47ResMLP-12
87
Percentage correct· Extra Data· 2021-05-07
ResMLP: Feedforward networks for image classification with data-efficient training Code
#48WRN-40-10, S=4
86.9
Percentage correct· Extra Data· 2020-11-30
Towards Better Accuracy-efficiency Trade-offs: Divide and Co-training Code
#49ResNet50 (A1)
86.9
Percentage correct· Extra Data· 2021-10-01
ResNet strikes back: An improved training procedure in timm Code
#50WRN-28-10 * 3
86.81
Percentage correct· Extra Data· 2021-03-10
MixMo: Mixing Multiple Inputs for Multiple Outputs via Deep Subnetworks Code
#51PyramidNet + AA (AMP)
86.64
Percentage correct· Extra Data· 2020-10-10
Regularizing Neural Networks via Adversarial Model Perturbation Code
#52PyramidNet-200 + Shakedrop + Cutmix + PS-KD
86.41
Percentage correct· Extra Data· 2020-06-22
Self-Knowledge Distillation with Progressive Refinement of Targets Code
#53Mixer-B/16- SAM
86.4
Percentage correct· Extra Data· 2021-06-03
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations Code
#54ResCNet-50
86.31
Percentage correct· 2024-11-16
Deep Feature Response Discriminative Calibration Code
#55PyramidNet-200 + Shakedrop + Cutmix
86.19
Percentage correct· Extra Data· 2019-05-13
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features Code
#56MUXNet-m
86.1
Percentage correct· Extra Data· 2020-03-31
MUXConv: Information Multiplexing in Convolutional Neural Networks Code
#57NAT-M1
86
Percentage correct· Extra Data· 2020-05-12
Neural Architecture Transfer Code
#58WRN-28-10
85.77
Percentage correct· Extra Data· 2021-03-10
MixMo: Mixing Multiple Inputs for Multiple Outputs via Deep Subnetworks Code
#59WRN-28-10, S=4
85.74
Percentage correct· Extra Data· 2020-11-30
Towards Better Accuracy-efficiency Trade-offs: Divide and Co-training Code
#60WRN-28-8 (SAMix+DM)
85.59
Percentage correct
No paper
#61WRN-28-8 +SAMix
85.5
Percentage correct· Extra Data· 2021-11-30
Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup Code
#62ASANas
85.42
Percentage correct· Extra Data· 2019-03-14
Improving Neural Architecture Search Image Classifiers via Ensemble Learning Code
#63WRN-28-8 (AutoMix+DM)
85.38
Percentage correct
No paper
#64SparseSwin
85.35
Percentage correct· Extra Data· 2023-09-11
SparseSwin: Swin Transformer with Sparse Transformer Block Code
#65WRN-28-8 (PuzzleMix+DM)
85.25
Percentage correct
No paper
#66ResNet-50-SAM
85.2
Percentage correct· Extra Data· 2021-06-03
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations Code
#67WRN-28-8 +AutoMix
85.16
Percentage correct· Extra Data· 2021-03-24
AutoMix: Unveiling the Power of Mixup for Stronger Classifiers Code
#68WaveMixLite-256/7
85.09
Percentage correct· Extra Data· 2022-05-28
WaveMix: A Resource-efficient Neural Network for Image Analysis Code
#69MANO-tiny
85.08
Percentage correct· Extra Data· 2025-07-03
Linear Attention with Global Context: A Multipole Attention Mechanism for Vision and Physics Code
#70WRN 28-14
85
Percentage correct· Extra Data· 2020-07-25
Neural networks with late-phase weights Code
#71R-Mix (WideResNet 28-10)
85
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#72EEEA-Net-C (b=5)+ CO
84.98
Percentage correct· Extra Data· 2021-08-13
EEEA-Net: An Early Exit Evolutionary Neural Architecture Search Code
#73RL-Mix (WideResNet 28-10)
84.9
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#74Wide-ResNet-28-10
84.89
Percentage correct· Extra Data· 2022-09-29
Automatic Data Augmentation via Invariance-Constrained Learning Code
#75SENet + ShakeEven + CutoutSOTA
84.59
Percentage correct· Extra Data· 2017-09-05
Squeeze-and-Excitation Networks Code
#76ResNeXt-50(32x4d) + SAMix
84.42
Percentage correct· Extra Data· 2021-11-30
Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup Code
#77WRN-28-10 with reSGHMC
84.38
Percentage correct· Extra Data· 2020-08-12
Non-convex Learning via Replica Exchange Stochastic Gradient MCMC Code
#78PyramidNet-272 + SWA
84.16
Percentage correct· Extra Data· 2018-03-14
Averaging Weights Leads to Wider Optima and Better Generalization Code
#79WRN28-10
84.05
Percentage correct· Extra Data· 2020-09-15
Puzzle Mix: Exploiting Saliency and Local Statistics for Optimal Mixup Code
#80HCGNet-A3
84.04
Percentage correct· Extra Data· 2019-08-26
Gated Convolutional Networks with Hybrid Connectivity for Image Classification Code
#81WideResNet 28-10 + CutMix (OneCycleLR scheduler)
83.97
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#82DenseNet-BC-190 + FMix
83.95
Percentage correct· Extra Data· 2020-02-27
FMix: Enhancing Mixed Sample Data Augmentation Code
#83ORNSOTA
83.85
Percentage correct· Extra Data· 2017-01-07
Oriented Response Networks Code
#84Grafit (ResNet-50)
83.7
Percentage correct· Extra Data· 2020-11-25
Grafit: Learning fine-grained image representations with coarse labels
#85ResNeXt-50(32x4d) + AutoMix
83.64
Percentage correct· Extra Data· 2021-03-24
AutoMix: Unveiling the Power of Mixup for Stronger Classifiers Code
#86CCT-7/3x1+HTM+VTM
83.57
Percentage correct· Extra Data· 2022-10-14
TokenMixup: Efficient Attention-guided Token-level Data Augmentation for Transformers Code
#87HCGNet-A2
83.46
Percentage correct· Extra Data· 2019-08-26
Gated Convolutional Networks with Hybrid Connectivity for Image Classification Code
#88Res2NeXt-29
83.44
Percentage correct· Extra Data· 2019-04-02
Res2Net: A New Multi-scale Backbone Architecture Code
#89DenseNet-BC-190 + Mixup
83.2
Percentage correct· Extra Data· 2017-10-25
mixup: Beyond Empirical Risk Minimization Code
#90SSAL-DenseNet 190-40
83.2
Percentage correct· Extra Data· 2021-01-07
Contextual Classification Using Self-Supervised Auxiliary Models for Deep Neural Networks Code
#91EnAET
83.13
Percentage correct· Extra Data· 2019-11-21
EnAET: A Self-Trained framework for Semi-Supervised and Supervised Learning with Ensemble Transformations Code
#92WRN 28-10
83.06
Percentage correct· Extra Data· 2020-07-25
Neural networks with late-phase weights Code
#93R-Mix (ResNeXt 29-4-24)
83.02
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#94Wide ResNet+Cutout+no BN scale/offset learning
82.95
Percentage correct· Extra Data· 2019-07-16
Single-bit-per-weight deep convolutional neural networks without batch-normalization layers for embedded systems Code
#95WRN-16-8 with reSGHMC
82.95
Percentage correct· Extra Data· 2020-08-12
Non-convex Learning via Replica Exchange Stochastic Gradient MCMC Code
#96DenseNet-BCSOTA
82.82
Percentage correct· Extra Data· 2016-08-25
Densely Connected Convolutional Networks Code
#97ABNet-2G-R3-Combined
82.784
Percentage correct· 2024-11-28
ANDHRA Bandersnatch: Training Neural Networks to Predict Parallel Realities Code
#98CCT-7/3x1*
82.72
Percentage correct· Extra Data· 2021-04-12
Escaping the Big Data Paradigm with Compact Transformers Code
#99EXACT (WRN-28-10)
82.68
Percentage correct· 2022-05-19
EXACT: How to Train Your Accuracy Code
#100SKNet-29 (ResNeXt-29, 16×32d)
82.67
Percentage correct· Extra Data· 2019-03-15
Selective Kernel Networks Code
#101DenseNet
82.62
Percentage correct· Extra Data· 2016-08-25
Densely Connected Convolutional Networks Code
#102Shared WRN
82.57
Percentage correct· Extra Data· 2019-02-26
Learning Implicitly Recurrent CNNs Through Parameter Sharing Code
#103Transformer local-attention (NesT-B)
82.56
Percentage correct· Extra Data· 2021-05-26
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding Code
#104RL-Mix (ResNeXt 29-4-24)
82.43
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#105Mixer-S/16- SAM
82.4
Percentage correct· Extra Data· 2021-06-03
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations Code
#106R-Mix (WideResNet 16-8)
82.32
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#107ResNeXt 29-4-24 + CutMix (OneCycleLR scheduler)
82.3
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#108WARN
82.18
Percentage correct· Extra Data· 2018-07-19
Attend and Rectify: a Gated Attention Mechanism for Fine-Grained Recovery Code
#109RL-Mix (WideResNet 16-8)
82.16
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#110WRN+SWA
82.15
Percentage correct· Extra Data· 2018-03-14
Averaging Weights Leads to Wider Optima and Better Generalization Code
#111Manifold Mixup
81.96
Percentage correct· Extra Data· 2018-06-13
Manifold Mixup: Better Representations by Interpolating Hidden States Code
#112HCGNet-A1
81.87
Percentage correct· Extra Data· 2019-08-26
Gated Convolutional Networks with Hybrid Connectivity for Image Classification Code
#113WideResNet 16-8 + CutMix (OneCycleLR scheduler)
81.79
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#114Residual Gates + WRN
81.73
Percentage correct· Extra Data· 2016-11-04
Learning Identity Mappings with Residual Gates
#115kNN-CLIP
81.7
Percentage correct· Extra Data· 2022-04-03
Revisiting a kNN-based Image Classification System with High-capacity Storage
#116AA-Wide-ResNet
81.6
Percentage correct· Extra Data· 2019-04-22
Attention Augmented Convolutional Networks Code
#117PDO-eConv (p8, 4.6M)
81.6
Percentage correct· Extra Data· 2020-07-20
PDO-eConvs: Partial Differential Operator Based Equivariant Convolutions Code
#118SEER (RegNet10B)
81.53
Percentage correct· Extra Data· 2022-02-16
Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision Code
#119R-Mix (PreActResNet-18)
81.49
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#120ResNet50 (FSGDM)
81.44
Percentage correct· 2024-11-29
On the Performance Analysis of Momentum Method: A Frequency Domain Perspective Code
#121Wide-ResNet-40-2
81.19
Percentage correct· Extra Data· 2022-09-29
Automatic Data Augmentation via Invariance-Constrained Learning Code
#122Wide ResNetSOTA
81.15
Percentage correct· Extra Data· 2016-05-23
Wide Residual Networks Code
#123CoPaNet-R-164
81.1
Percentage correct· Extra Data· 2017-09-29
Deep Competitive Pathway Networks Code
#124ABNet-2G-R3
80.83
Percentage correct· 2024-11-28
ANDHRA Bandersnatch: Training Neural Networks to Predict Parallel Realities Code
#125RL-Mix (PreActResNet-18)
80.75
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#126PreActResNet-18 + CutMix (OneCycleLR scheduler)
80.6
Percentage correct· Extra Data· 2022-12-09
Expeditious Saliency-guided Mix-up through Random Gradient Thresholding Code
#127GAC-SNN
80.45
Percentage correct· 2023-08-12
Gated Attention Coding for Training High-performance and Efficient Spiking Neural Networks Code
#128ABNet-2G-R2
80.354
Percentage correct· 2024-11-28
ANDHRA Bandersnatch: Training Neural Networks to Predict Parallel Realities Code
#129SimpleNetv2
80.29
Percentage correct· Extra Data· 2018-02-17
Towards Principled Design of Deep Convolutional Networks: Introducing SimpNet Code
#130UPANets
80.29
Percentage correct· Extra Data· 2021-03-15
UPANets: Learning from the Universal Pixel Attention Networks Code
#131PreActResNet-18 + SageMix
80.16
Percentage correct· Extra Data· 2022-10-13
SageMix: Saliency-Guided Mixup for Point Clouds Code
#132ResNet56 with reSGHMC
80.14
Percentage correct· Extra Data· 2020-08-12
Non-convex Learning via Replica Exchange Stochastic Gradient MCMC Code
#133PDO-eConv (p8, 2.62M)
79.99
Percentage correct· Extra Data· 2020-07-20
PDO-eConvs: Partial Differential Operator Based Equivariant Convolutions Code
#134VGG11B(3x) + LocalLearning
79.9
Percentage correct· Extra Data· 2019-01-20
Training Neural Networks with Local Error Signals Code
#135NNCLR
79
Percentage correct· Extra Data· 2021-04-29
With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representations Code
#136ABNet-2G-R1
78.792
Percentage correct· 2024-11-28
ANDHRA Bandersnatch: Training Neural Networks to Predict Parallel Realities Code
#137PreActResNet18 (AMP)
78.49
Percentage correct· Extra Data· 2020-10-10
Regularizing Neural Networks via Adversarial Model Perturbation Code
#138SimpleNetv1
78.37
Percentage correct· Extra Data· 2016-08-22
Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures Code
#139ViT (lightweight, MAE pre-trained)
78.27
Percentage correct· 2024-02-06
Pre-training of Lightweight Vision Transformers on Small Datasets with Minimally Scaled Images
#140PDC
77.9
Percentage correct· Extra Data· 2021-04-16
Augmenting Deep Classifiers with Polynomial Neural Networks Code
#141MobileNetV3-large x1.0 (BSConv-U)
77.7
Percentage correct· Extra Data· 2020-03-30
Rethinking Depthwise Separable Convolutions: How Intra-Kernel Correlations Lead to Improved MobileNets Code
#142CCT-6/3x1
77.31
Percentage correct· Extra Data· 2021-04-12
Escaping the Big Data Paradigm with Compact Transformers Code
#143ResNet-1001SOTA
77.3
Percentage correct· Extra Data· 2016-03-16
Identity Mappings in Deep Residual Networks Code
#144Evolution
77
Percentage correct· Extra Data· 2017-03-03
Large-Scale Evolution of Image Classifiers Code
#145DIANet
76.98
Percentage correct· Extra Data· 2019-05-25
DIANet: Dense-and-Implicit Attention Network Code
#146LP-BNN (ours) + cutout
76.85
Percentage correct· Extra Data· 2020-12-04
Encoding the latent posterior of Bayesian Neural Networks for uncertainty quantification Code
#147ResNet-18+MM+FRL
76.64
Percentage correct· Extra Data· 2020-11-22
Learning Class Unique Features in Fine-Grained Visual Classification
#148ResNet32 with reSGHMC
76.55
Percentage correct· Extra Data· 2020-08-12
Non-convex Learning via Replica Exchange Stochastic Gradient MCMC Code
#149MomentumNet
76.38
Percentage correct· Extra Data· 2021-02-15
Momentum Residual Neural Networks Code
#150SSCNNSOTA
75.7
Percentage correct· Extra Data· 2014-09-22
Spatially-sparse convolutional neural networks Code
#151Exponential Linear Units
75.7
Percentage correct· Extra Data· 2015-11-23
Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)Code
#152ResNet-9
75.59
Percentage correct· Extra Data· 2022-03-29
CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters Code
#153Stochastic Depth
75.42
Percentage correct· Extra Data· 2016-03-30
Deep Networks with Stochastic Depth Code
#154ResNet v2-110 (Mish activation)
74.41
Percentage correct· Extra Data· 2019-08-23
Mish: A Self Regularized Non-Monotonic Activation Function Code
#155Dspike (ResNet-18)
74.24
Percentage correct
No paper
#156ResNet20 with reSGHMC
74.14
Percentage correct· Extra Data· 2020-08-12
Non-convex Learning via Replica Exchange Stochastic Gradient MCMC Code
#157MixMatch
74.1
Percentage correct· Extra Data· 2019-05-06
MixMatch: A Holistic Approach to Semi-Supervised Learning Code
#158Beta-Rank
74.01
Percentage correct· 2023-04-15
Beta-Rank: A Robust Convolutional Filter Pruning Method For Imbalanced Medical Image Analysis Code
#159PreResNet-110
73.98
Percentage correct· 2023-02-13
How to Use Dropout Correctly on Residual Networks with Batch Normalization Code
#160ABNet-2G-R0
73.93
Percentage correct· 2024-11-28
ANDHRA Bandersnatch: Training Neural Networks to Predict Parallel Realities Code
#161Fractional MP
73.6
Percentage correct· Extra Data· 2014-12-18
Fractional Max-Pooling Code
#162ResNet+ELU
73.5
Percentage correct· 2016-04-14
Deep Residual Networks with Exponential Linear Unit Code
#163PDO-eConv (p6m,0.37M)
73
Percentage correct· Extra Data· 2020-07-20
PDO-eConvs: Partial Differential Operator Based Equivariant Convolutions Code
#164SOPCNN
72.96
Percentage correct· Extra Data· 2020-01-24
Stochastic Optimization of Plain Convolutional Neural Networks with Simple methods Code
#165PDO-eConv (p6,0.36M)
72.87
Percentage correct· Extra Data· 2020-07-20
PDO-eConvs: Partial Differential Operator Based Equivariant Convolutions Code
#166Tuned CNN
72.6
Percentage correct· Extra Data· 2015-02-19
Scalable Bayesian Optimization Using Deep Neural Networks Code
#167ResNet-110 (SAP)
72.537
Percentage correct· 2024-09-25
Stochastic Subsampling With Average Pooling
#168CMsC
72.4
Percentage correct· Extra Data· 2015-11-18
Competitive Multi-scale Convolution
#169Fitnet4-LSUV
72.3
Percentage correct· Extra Data· 2015-11-19
All you need is a good init Code
#170GAN+ResNet
71.52
Percentage correct
No paperCode
#171kMobileNet V3 Large 16ch
71.36
Percentage correct· Extra Data
No paperCode
#172BNM NiN
71.1
Percentage correct· Extra Data· 2015-11-09
Batch-normalized Maxout Network in Network Code
#173OTTT
71.05
Percentage correct· 2022-10-09
Online Training Through Time for Spiking Neural Networks Code
#174MIM
70.8
Percentage correct· Extra Data· 2015-08-03
On the Importance of Normalisation Layers in Deep Learning with Piecewise Linear Activation Units
#175WaveMix-Lite-256/7
70.2
Percentage correct· 2022-05-28
WaveMix: A Resource-efficient Neural Network for Image Analysis Code
#176IM-Loss (VGG-16)
70.18
Percentage correct
No paper
#177NiN+APL
69.2
Percentage correct· Extra Data· 2014-12-21
Learning Activation Functions to Improve Deep Neural Networks Code
#178SWWAE
69.1
Percentage correct· Extra Data· 2015-06-08
Stacked What-Where Auto-encoders Code
#179NiN+Superclass+CDJ
69
Percentage correct· Extra Data· 2017-06-06
Deep Convolutional Decision Jungle for Image Classification
#180Spectral Representations for Convolutional Neural Networks
68.4
Percentage correct· 2015-06-11
Spectral Representations for Convolutional Neural Networks
#181ReActNet-18
68.34
Percentage correct· 2021-04-16
"BNN - BN = ?": Training Binary Neural Networks without Batch Normalization Code
#182VDN
67.8
Percentage correct· 2015-07-22
Training Very Deep Networks Code
#183DCNN+GFE
67.7
Percentage correct· 2017-10-06
Deep Convolutional Neural Networks as Generic Feature Extractors
#184Tree+Max-Avg pooling
67.6
Percentage correct· 2015-09-30
Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree Code
#185HD-CNN
67.4
Percentage correct· 2014-10-03
HD-CNN: Hierarchical Deep Convolutional Neural Network for Large Scale Visual Recognition Code
#186Universum Prescription
67.2
Percentage correct· 2015-11-11
Universum Prescription: Regularization using Unlabeled Data
#187ResNet50 Without Transfer Learning
67.06
Percentage correct
No paperCode
#188AlexNet (KP)
66.78
Percentage correct
No paper
#189ACN
66.3
Percentage correct· 2014-12-21
Striving for Simplicity: The All Convolutional Net Code
#190DLME (ResNet-18, linear)
66.1
Percentage correct· 2022-07-07
DLME: Deep Local-flatness Manifold Embedding Code
#191ResNet-18 (modified)
66
Percentage correct· 2022-10-30
FatNet: High Resolution Kernels for Classification Using Fully Convolutional Optical Neural Networks Code
#192DSNSOTA
65.4
Percentage correct· 2014-09-18
Deeply-Supervised Nets Code
#193NiNSOTA
64.3
Percentage correct· 2013-12-16
Network In Network Code
#194Tree Priors
63.2
Percentage correct
No paper
#195DNN+Probabilistic Maxout
61.9
Percentage correct· 2013-12-20
Improving Deep Neural Networks with Probabilistic Maxout Units
#196Maxout Network (k=2)SOTA
61.43
Percentage correct· 2013-02-18
Maxout Networks Code
#197ResNet20+UnsharpMaskLayer
60.36
Percentage correct
No paperCode
#198Convolutional Linear Transformer for Vision (CLTV)
60.11
Percentage correct· 2022-01-25
Convolutional Xformers for Vision Code
#199FatNet of ResNet-18
60
Percentage correct· 2022-10-30
FatNet: High Resolution Kernels for Classification Using Fully Convolutional Optical Neural Networks Code
#200Optical Simulation of FatNet
60
Percentage correct· 2022-10-30
FatNet: High Resolution Kernels for Classification Using Fully Convolutional Optical Neural Networks Code
#201RReLU
59.8
Percentage correct· 2015-05-05
Empirical Evaluation of Rectified Activations in Convolutional Network Code
#202Stochastic PoolingSOTA
57.5
Percentage correct· 2013-01-16
Stochastic Pooling for Regularization of Deep Convolutional Neural Networks Code
#203Sign-symmetry
48.75
Percentage correct· 2015-10-17
How Important is Weight Symmetry in Backpropagation?Code
#204AlexNet (DFA)
48.03
Percentage correct
No paper
#205CNN39
42.64
Percentage correct· 2020-10-03
Sharpness-Aware Minimization for Efficiently Improving Generalization Code
#206CNN36
36.07
Percentage correct· 2020-10-03
Sharpness-Aware Minimization for Efficiently Improving Generalization Code
#207CNN37
35.05
Percentage correct· 2021-11-24
Sharpness-aware Quantization for Deep Neural Networks Code
#208AlexNet (FA)
19.49
Percentage correct
No paper