You Don't Need Domain-Specific Data Augmentations When Scaling Self-Supervised Learning

• URL: http://arxiv.org/abs/2406.09294v2
• Date: Fri, 29 Nov 2024 13:39:02 GMT
• Title: You Don't Need Domain-Specific Data Augmentations When Scaling Self-Supervised Learning
• Authors: Théo Moutakanni, Maxime Oquab, Marc Szafraniec, Maria Vakalopoulou, Piotr Bojanowski,
• Abstract summary: Self-Supervised learning (SSL) with Joint-Embedding Architectures (JEA) has led to outstanding performances. generative reconstruction-based models have shown strong performance without using data augmentations except masking. We show that strong image representations can be obtained with JEAs and only cropping without resizing provided the training data is large enough.
• Score: 8.384940156285847
• License:
• Abstract: Self-Supervised learning (SSL) with Joint-Embedding Architectures (JEA) has led to outstanding performances. All instantiations of this paradigm were trained using strong and well-established hand-crafted data augmentations, leading to the general belief that they are required for the proper training and performance of such models. On the other hand, generative reconstruction-based models such as BEIT and MAE or Joint-Embedding Predictive Architectures such as I-JEPA have shown strong performance without using data augmentations except masking. In this work, we challenge the importance of invariance and data-augmentation in JEAs at scale. By running a case-study on a recent SSL foundation model - DINOv2 - we show that strong image representations can be obtained with JEAs and only cropping without resizing provided the training data is large enough, reaching state-of-the-art results and using the least amount of augmentation in the literature. Through this study, we also discuss the impact of compute constraints on the outcomes of experimental deep learning research, showing that they can lead to very different conclusions.

Related papers

• SMPLest-X: Ultimate Scaling for Expressive Human Pose and Shape Estimation [81.36747103102459]
  Expressive human pose and shape estimation (EHPS) unifies body, hands, and face motion capture with numerous applications. Current state-of-the-art methods focus on training innovative architectural designs on confined datasets. We investigate the impact of scaling up EHPS towards a family of generalist foundation models.
  arXiv  Detail & Related papers  (2025-01-16T18:59:46Z)
• T-JEPA: Augmentation-Free Self-Supervised Learning for Tabular Data [0.0]
  Self-supervised learning (SSL) generally involves generating different views of the same sample and thus requires data augmentations. In the present work, we propose a novel augmentation-free SSL method for structured data. Our approach, T-JEPA, relies on a Joint Embedding Predictive Architecture (JEPA) and is akin to mask reconstruction in the latent space.
  arXiv  Detail & Related papers  (2024-10-07T13:15:07Z)
• Efficient Training of Self-Supervised Speech Foundation Models on a Compute Budget [53.882492597728195]
  This paper investigates how to efficiently train speech foundation models with self-supervised learning (SSL) under a limited compute budget. We examine critical factors in SSL that impact the budget, including model architecture, model size, and data size.
  arXiv  Detail & Related papers  (2024-09-09T10:36:42Z)
• Fantastic Gains and Where to Find Them: On the Existence and Prospect of General Knowledge Transfer between Any Pretrained Model [74.62272538148245]
  We show that for arbitrary pairings of pretrained models, one model extracts significant data context unavailable in the other. We investigate if it is possible to transfer such "complementary" knowledge from one model to another without performance degradation.
  arXiv  Detail & Related papers  (2023-10-26T17:59:46Z)
• Exploring the Impact of Instruction Data Scaling on Large Language Models: An Empirical Study on Real-World Use Cases [17.431381376675432]
  In this paper we explore the performance of large language models based on instruction tuning across different scales of instruction data. With Bloomz-7B1-mt as the base model, the results show that merely increasing the amount of instruction data leads to continuous improvement in tasks such as open-ended generation. We propose potential future research directions such as effectively selecting high-quality training data, scaling base models and training methods specialized for hard tasks.
  arXiv  Detail & Related papers  (2023-03-26T14:49:37Z)
• Instance-Conditioned GAN Data Augmentation for Representation Learning [29.36473147430433]
  We introduce DA_IC-GAN, a learnable data augmentation module that can be used off-the-shelf in conjunction with most state-of-the-art training recipes. We show that DA_IC-GAN can boost accuracy to between 1%p and 2%p with the highest capacity models. We additionally couple DA_IC-GAN with a self-supervised training recipe and show that we can also achieve an improvement of 1%p in accuracy in some settings.
  arXiv  Detail & Related papers  (2023-03-16T22:45:43Z)
• Exploring the Effects of Data Augmentation for Drivable Area Segmentation [0.0]
  We focus on investigating the benefits of data augmentation by analyzing pre-existing image datasets. Our results show that the performance and robustness of existing state of the art (or SOTA) models can be increased dramatically.
  arXiv  Detail & Related papers  (2022-08-06T03:39:37Z)
• Data augmentation for learning predictive models on EEG: a systematic comparison [79.84079335042456]
  deep learning for electroencephalography (EEG) classification tasks has been rapidly growing in the last years. Deep learning for EEG classification tasks has been limited by the relatively small size of EEG datasets. Data augmentation has been a key ingredient to obtain state-of-the-art performances across applications such as computer vision or speech.
  arXiv  Detail & Related papers  (2022-06-29T09:18:15Z)
• Models, Pixels, and Rewards: Evaluating Design Trade-offs in Visual Model-Based Reinforcement Learning [109.74041512359476]
  We study a number of design decisions for the predictive model in visual MBRL algorithms. We find that a range of design decisions that are often considered crucial, such as the use of latent spaces, have little effect on task performance. We show how this phenomenon is related to exploration and how some of the lower-scoring models on standard benchmarks will perform the same as the best-performing models when trained on the same training data.
  arXiv  Detail & Related papers  (2020-12-08T18:03:21Z)
• Relation-Guided Representation Learning [53.60351496449232]
  We propose a new representation learning method that explicitly models and leverages sample relations. Our framework well preserves the relations between samples. By seeking to embed samples into subspace, we show that our method can address the large-scale and out-of-sample problem.
  arXiv  Detail & Related papers  (2020-07-11T10:57:45Z)

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.