| Efficient Global Neural Architecture Search | ✓ Link | 99.35 | | 420000 | KMNIST-Tiny | 2025-02-08 |
| Efficient Global Neural Architecture Search | ✓ Link | 99.29 | | 2710000 | KMNIST-Mobile | 2025-02-08 |
| SpinalNet: Deep Neural Network with Gradual Input | ✓ Link | 99.15 | 0.85 | | VGG-5 (Spinal FC) | 2020-07-07 |
| Context-Aware Multipath Networks | | 99.05 | 0.95 | | CAMNet3 | 2019-07-26 |
| Training Neural Networks with Local Error Signals | ✓ Link | 99.01 | 0.99 | | VGG8B(2x) + LocalLearning + CO | 2019-01-20 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.84 | | | CN(d=32) | 2021-01-01 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.81 | | | NSRL (log D) (d=16) | 2021-01-01 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.80 | | | CN(d=16) | 2021-01-01 |
| A Comprehensive Study of ImageNet Pre-Training for Historical Document Image Analysis | | 98.79 | | | Resnet-152 | 2019-05-22 |
| Learning local discrete features in explainable-by-design convolutional neural networks | ✓ Link | 98.78 | 1.22 | 892362 | R-ExplaiNet-26 | 2024-10-31 |
| CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters | ✓ Link | 98.75 | | | ResNet-14 | 2022-03-29 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.72 | | | NSRL (WGAN) (d=32) | 2021-01-01 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.68 | | | NSRL (WGAN) (d=8) | 2021-01-01 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.66 | | | NSRL (WGAN) (d=16) | 2021-01-01 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.63 | | | NSRL (log D) (d=32) | 2021-01-01 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.61 | | | NSRL (log D) (d=8) | 2021-01-01 |
| Toward Understanding Supervised Representation Learning with RKHS and GAN | | 98.60 | | | CN(d=8) | 2021-01-01 |
| Improved efficient capsule network for Kuzushiji-MNIST benchmark dataset classification | ✓ Link | 98.43 | | | Efficient Capsnet | 2023-12-15 |
| mixup: Beyond Empirical Risk Minimization | ✓ Link | 98.41 | | | PreActResNet-18 + Input Mixup | 2017-10-25 |
| Identity Mappings in Deep Residual Networks | ✓ Link | 97.82 | | | PreActResNet-18 | 2016-03-16 |
| The Convolutional Tsetlin Machine | ✓ Link | 96.3 | | | Convolutional Tsetlin Machine | 2019-05-23 |
| KerCNNs: biologically inspired lateral connections for classification of corrupted images | | 93.13 | | | KerCNN | 2019-10-18 |
| Multi-Complementary and Unlabeled Learning for Arbitrary Losses and Models | | 79.90 | | | linear/flexible model | 2020-01-13 |
| Multi-Complementary and Unlabeled Learning for Arbitrary Losses and Models | | 79.5 | | | FWD | 2020-01-13 |
| Complementary-Label Learning for Arbitrary Losses and Models | ✓ Link | 67.1 | | | Complementary-Label Learning | 2018-10-10 |
| Deep Learning for Classical Japanese Literature | ✓ Link | | 1.10 | | ResNet18 + VGG Ensemble | 2018-12-03 |