AI Paper Atlas

Research map / Benchmarks / ImageNet, ResNet-18

ImageNet, ResNet-18 leaderboard

31 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ImageNet, ResNet-18. Best accuracy: 73.1 (DCD, ICLR 2021).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

ImageNet, ResNet-18 · accuracy

31 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
173.1DCD
ResNet-18
Revisiting Dynamic Convolution via Matrix Decomposition
ICLR 2021 · Table: Comparing DCD with vani
CNN design
271.83NIPQ-MP-4+KD
ResNet-18
NIPQ: Noise Proxy-Based Integrated Pseudo-Quantization
CVPR 2023 · Table: ImageNet classification
Quantization
371.56HAWQ-V3 INT8
ResNet18
HAWQ-V3: Dyadic Neural Network Quantization
ICML 2021 · Table 1
Quantization
471.3FST
ResNet-18
Towards Fully Sparse Training: Information Restoration with Spatial Similarity
AAAI 2022 · Table 3
Pruning and efficient tuning
571.3DDQ, 4-bit weights/8-bit activations, 90 epochs
ResNet18
Differentiable Dynamic Quantization with Mixed Precision and Adaptive Resolution
ICML 2021 · Table 1
Quantization
671.23MCA-E
ResNet-18
MCA: Moment Channel Attention Networks
AAAI 2024 · Table: Image classification re
CNN features
771Ours, 4MP/4MP
ResNet18
Retraining-Free Model Quantization via One-Shot Weight-Coupling Learning
CVPR 2024 · Table 2
Quantization
870.9APoT 5/5
ResNet-18
Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networks
ICLR 2020 · Table 1
Quantization
970.58SGDHess
ResNet18
Better SGD using Second-order Momentum
NeurIPS 2022 · Table, Top 1 accuracy on Image
Optimizers
1070.48MMEL-H
ResNet18
Reweighting Augmented Samples by Minimizing the Maximal Expected Loss
ICLR 2021 · ImageNet table
Recognition and long-tail
21 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.