Simplifying DINO via Coding Rate Regularization
This work simplifies and improves the DINO and DINOv2 models, which are widely used for learning representations from unlabeled imagery data, making them more accessible and effective for downstream tasks such as image classification and segmentation.
The authors simplified the DINO and DINOv2 models by removing empirically motivated design choices and adding a coding rate term to the loss function, resulting in more robust models with higher-quality representations. The simplified models, SimDINO and SimDINOv2, offer a Pareto improvement over the original models.
DINO and DINOv2 are two model families being widely used to learn representations from unlabeled imagery data at large scales. Their learned representations often enable state-of-the-art performance for downstream tasks, such as image classification and segmentation. However, they employ many empirically motivated design choices and their training pipelines are highly complex and unstable -- many hyperparameters need to be carefully tuned to ensure that the representations do not collapse -- which poses considerable difficulty to improving them or adapting them to new domains. In this work, we posit that we can remove most such-motivated idiosyncrasies in the pre-training pipelines, and only need to add an explicit coding rate term in the loss function to avoid collapse of the representations. As a result, we obtain highly simplified variants of the DINO and DINOv2 which we call SimDINO and SimDINOv2, respectively. Remarkably, these simplified models are more robust to different design choices, such as network architecture and hyperparameters, and they learn even higher-quality representations, measured by performance on downstream tasks, offering a Pareto improvement over the corresponding DINO and DINOv2 models. This work highlights the potential of using simplifying design principles to improve the empirical practice of deep learning.