Command Palette
Search for a command to run...
H M Dipu Kabir
Abstract
Multitask learning is a popular approach to training high-performing neural networks with improved generalization. In this paper, we propose a background class to achieve improved generalization at a lower computation compared to multitask learning to help researchers and organizations with limited computation power. We also present a methodology for selecting background images and discuss potential future improvements. We apply our approach to several datasets and achieve improved generalization with much lower computation. Through the class activation mappings (CAMs) of the trained models, we observed the tendency towards looking at a bigger picture with the proposed model training methodology. Applying the vision transformer with the proposed background class, we receive state-of-the-art (SOTA) performance on CIFAR-10C, Caltech-101, and CINIC-10 datasets. Example scripts are available in the `CAM' folder of the following GitHub Repository: github.com/dipuk0506/UQ
Code Repositories
Benchmarks
| Benchmark | Methodology | Metrics |
|---|---|---|
| classification-on-cifar-10c | ViT-L/16 (Background) | Accuracy on Brightness Corrupted Images: 99.03 |
| fine-grained-image-classification-on-caltech | VIT-L/16 | Top-1 Error Rate: 1.98% |
| image-classification-on-cifar-10 | VIT-L/16 (Spinal FC, Background) | Percentage correct: 99.05 |
| image-classification-on-cifar-100 | VIT-L/16 (Spinal FC, Background) | Percentage correct: 93.31 |
| image-classification-on-cinic-10 | VIT-L/16 (Spinal FC, Background) | Accuracy: 95.80 |
| image-classification-on-flowers-102 | VIT-L/16 (Background) | Accuracy: 99.75 |
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.