On Partial Prototype Collapse in the DINO Family of Self-Supervised Methods

• URL: http://arxiv.org/abs/2410.14060v1
• Date: Thu, 17 Oct 2024 22:06:34 GMT
• Title: On Partial Prototype Collapse in the DINO Family of Self-Supervised Methods
• Authors: Hariprasath Govindarajan, Per Sidén, Jacob Roll, Fredrik Lindsten,
• Abstract summary: Learning to map the data samples to compact representations leads to the representation collapse problem. Regularizing the distribution of data points over the clusters is the prevalent strategy to avoid this issue. We show that a partial prototype collapse problem still exists in the DINO family of methods, that leads to significant redundancies in the prototypes.
• Score: 15.524425102344784
• License:
• Abstract: A prominent self-supervised learning paradigm is to model the representations as clusters, or more generally as a mixture model. Learning to map the data samples to compact representations and fitting the mixture model simultaneously leads to the representation collapse problem. Regularizing the distribution of data points over the clusters is the prevalent strategy to avoid this issue. While this is sufficient to prevent full representation collapse, we show that a partial prototype collapse problem still exists in the DINO family of methods, that leads to significant redundancies in the prototypes. Such prototype redundancies serve as shortcuts for the method to achieve a marginal latent class distribution that matches the prescribed prior. We show that by encouraging the model to use diverse prototypes, the partial prototype collapse can be mitigated. Effective utilization of the prototypes enables the methods to learn more fine-grained clusters, encouraging more informative representations. We demonstrate that this is especially beneficial when pre-training on a long-tailed fine-grained dataset.

Related papers

• Data Pruning in Generative Diffusion Models [2.0111637969968]
  Generative models aim to estimate the underlying distribution of the data, so presumably they should benefit from larger datasets. We show that eliminating redundant or noisy data in large datasets is beneficial particularly when done strategically.
  arXiv  Detail & Related papers  (2024-11-19T14:13:25Z)
• Multi-Scale Grouped Prototypes for Interpretable Semantic Segmentation [7.372346036256517]
  Prototypical part learning is emerging as a promising approach for making semantic segmentation interpretable. We propose a method for interpretable semantic segmentation that leverages multi-scale image representation for prototypical part learning. Experiments conducted on Pascal VOC, Cityscapes, and ADE20K demonstrate that the proposed method increases model sparsity, improves interpretability over existing prototype-based methods, and narrows the performance gap with the non-interpretable counterpart models.
  arXiv  Detail & Related papers  (2024-09-14T17:52:59Z)
• Mixed Prototype Consistency Learning for Semi-supervised Medical Image Segmentation [0.0]
  We propose the Mixed Prototype Consistency Learning (MPCL) framework, which includes a Mean Teacher and an auxiliary network. The Mean Teacher generates prototypes for labeled and unlabeled data, while the auxiliary network produces additional prototypes for mixed data processed by CutMix. High-quality global prototypes for each class are formed by fusing two enhanced prototypes, optimizing the distribution of hidden embeddings used in consistency learning.
  arXiv  Detail & Related papers  (2024-04-16T16:51:12Z)
• Beyond Known Clusters: Probe New Prototypes for Efficient Generalized Class Discovery [23.359450657842686]
  Generalized Class Discovery (GCD) aims to dynamically assign labels to unlabelled data partially based on knowledge learned from labelled data. We propose an adaptive probing mechanism that introduces learnable potential prototypes to expand cluster prototypes. Our method surpasses the nearest competitor by a significant margin of 9.7% within the Stanford Cars dataset.
  arXiv  Detail & Related papers  (2024-04-13T12:41:40Z)
• RanPAC: Random Projections and Pre-trained Models for Continual Learning [59.07316955610658]
  Continual learning (CL) aims to learn different tasks (such as classification) in a non-stationary data stream without forgetting old ones. We propose a concise and effective approach for CL with pre-trained models.
  arXiv  Detail & Related papers  (2023-07-05T12:49:02Z)
• Unicom: Universal and Compact Representation Learning for Image Retrieval [65.96296089560421]
  We cluster the large-scale LAION400M into one million pseudo classes based on the joint textual and visual features extracted by the CLIP model. To alleviate such conflict, we randomly select partial inter-class prototypes to construct the margin-based softmax loss. Our method significantly outperforms state-of-the-art unsupervised and supervised image retrieval approaches on multiple benchmarks.
  arXiv  Detail & Related papers  (2023-04-12T14:25:52Z)
• Rethinking Semantic Segmentation: A Prototype View [126.59244185849838]
  We present a nonparametric semantic segmentation model based on non-learnable prototypes. Our framework yields compelling results over several datasets. We expect this work will provoke a rethink of the current de facto semantic segmentation model design.
  arXiv  Detail & Related papers  (2022-03-28T21:15:32Z)
• Dual Prototypical Contrastive Learning for Few-shot Semantic Segmentation [55.339405417090084]
  We propose a dual prototypical contrastive learning approach tailored to the few-shot semantic segmentation (FSS) task. The main idea is to encourage the prototypes more discriminative by increasing inter-class distance while reducing intra-class distance in prototype feature space. We demonstrate that the proposed dual contrastive learning approach outperforms state-of-the-art FSS methods on PASCAL-5i and COCO-20i datasets.
  arXiv  Detail & Related papers  (2021-11-09T08:14:50Z)
• Learning Class-level Prototypes for Few-shot Learning [24.65076873131432]
  Few-shot learning aims to recognize new categories using very few labeled samples. We propose a framework for few-shot classification, which can learn to generate preferable prototypes from few support data.
  arXiv  Detail & Related papers  (2021-08-25T06:33:52Z)
• Attentional Prototype Inference for Few-Shot Segmentation [128.45753577331422]
  We propose attentional prototype inference (API), a probabilistic latent variable framework for few-shot segmentation. We define a global latent variable to represent the prototype of each object category, which we model as a probabilistic distribution. We conduct extensive experiments on four benchmarks, where our proposal obtains at least competitive and often better performance than state-of-the-art prototype-based methods.
  arXiv  Detail & Related papers  (2021-05-14T06:58:44Z)
• Visualising Deep Network's Time-Series Representations [93.73198973454944]
  Despite the popularisation of machine learning models, more often than not they still operate as black boxes with no insight into what is happening inside the model. In this paper, a method that addresses that issue is proposed, with a focus on visualising multi-dimensional time-series data. Experiments on a high-frequency stock market dataset show that the method provides fast and discernible visualisations.
  arXiv  Detail & Related papers  (2021-03-12T09:53:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.