Mean Shift for Self-Supervised Learning
This addresses the need for simpler and effective self-supervised learning methods in computer vision, though it appears incremental as it builds on existing approaches like BYOL.
The paper tackles the problem of self-supervised learning by introducing a mean-shift algorithm that groups images without contrasting or strong cluster priors, achieving 72.4% on ImageNet linear evaluation with ResNet50 at 200 epochs and outperforming BYOL.
Most recent self-supervised learning (SSL) algorithms learn features by contrasting between instances of images or by clustering the images and then contrasting between the image clusters. We introduce a simple mean-shift algorithm that learns representations by grouping images together without contrasting between them or adopting much of prior on the structure of the clusters. We simply "shift" the embedding of each image to be close to the "mean" of its neighbors. Since in our setting, the closest neighbor is always another augmentation of the same image, our model will be identical to BYOL when using only one nearest neighbor instead of 5 as used in our experiments. Our model achieves 72.4% on ImageNet linear evaluation with ResNet50 at 200 epochs outperforming BYOL. Our code is available here: https://github.com/UMBCvision/MSF