Search Results for author:
Found 6 papers, 1 papers with code
Understanding the Effect of using Semantically Meaningful Tokens for Visual Representation Learning
Vision transformers have established a precedent of patchifying images into uniformly-sized chunks before processing.
Augmentations vs Algorithms: What Works in Self-Supervised Learning
We study the relative effects of data augmentations, pretraining algorithms, and model architectures in Self-Supervised Learning (SSL).
Disentangling the Effects of Data Augmentation and Format Transform in Self-Supervised Learning of Image Representations
We define augmentations in frequency space called Fourier Domain Augmentations (FDA) and show that training SSL models on a combination of these and image augmentations can improve the downstream classification accuracy by up to 1. 3% on ImageNet-1K.
Adapting Self-Supervised Representations to Multi-Domain Setups
Current state-of-the-art self-supervised approaches, are effective when trained on individual domains but show limited generalization on unseen domains.
Identifying Interpretable Subspaces in Image Representations
Although many existing approaches interpret features independently, we observe in state-of-the-art self-supervised and supervised models, that less than 20% of the representation space can be explained by individual features.
Measuring Self-Supervised Representation Quality for Downstream Classification using Discriminative Features
Fine-tuning with Q-Score regularization can boost the linear probing accuracy of SSL models by up to 5. 8% on ImageNet-100 and 3. 7% on ImageNet-1K compared to their baselines.
