{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/dino-as-a-von-mises-fisher-mixture-model-1","title":"DINO as a von Mises-Fisher mixture model","arxiv_id":"2405.10939","date":"2024-05-17","proceeding":"ICLR 2023 2","authors":["Hariprasath Govindarajan","Per Sidén","Jacob Roll","Fredrik Lindsten"],"abstract":"Self-distillation methods using Siamese networks are popular for self-supervised pre-training. DINO is one such method based on a cross-entropy loss between $K$-dimensional probability vectors, obtained by applying a softmax function to the dot product between representations and learnt prototypes. Given the fact that the learned representations are $L^2$-normalized, we show that DINO and its derivatives, such as iBOT, can be interpreted as a mixture model of von Mises-Fisher components. With this interpretation, DINO assumes equal precision for all components when the prototypes are also $L^2$-normalized. Using this insight we propose DINO-vMF, that adds appropriate normalization constants when computing the cluster assignment probabilities. Unlike DINO, DINO-vMF is stable also for the larger ViT-Base model with unnormalized prototypes. We show that the added flexibility of the mixture model is beneficial in terms of better image representations. The DINO-vMF pre-trained model consistently performs better than DINO on a range of downstream tasks. We obtain similar improvements for iBOT-vMF vs iBOT and thereby show the relevance of our proposed modification also for other methods derived from DINO.","url_abs":"https://arxiv.org/abs/2405.10939v1","url_pdf":"https://arxiv.org/pdf/2405.10939v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"model","task_name":"model"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"dino","method_name":"DINO"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/self-supervised-image-classification-on","task":"Self-Supervised Image Classification","dataset":"ImageNet","model":"iBOT-vMF (ViT-B/16)","rank_in_archive_order":24,"of":144,"metrics":{"Number of Params":"85M","Top 1 Accuracy":"80.3%"},"uses_additional_data":false},{"leaderboard":"/sota/self-supervised-image-classification-on","task":"Self-Supervised Image Classification","dataset":"ImageNet","model":"DINO-vMF (ViT-B/16)","rank_in_archive_order":42,"of":144,"metrics":{"Number of Params":"85M","Top 1 Accuracy":"78.8%"},"uses_additional_data":false},{"leaderboard":"/sota/self-supervised-image-classification-on","task":"Self-Supervised Image Classification","dataset":"ImageNet","model":"DINO-vMF (ViT-S/16)","rank_in_archive_order":57,"of":144,"metrics":{"Number of Params":"21M","Top 1 Accuracy":"77.0%"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2405.10939","atlas_url":"https://app.syntology.ai/?focus=2405.10939","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}