Infinite Class Mixup

• URL: http://arxiv.org/abs/2305.10293v2
• Date: Wed, 6 Sep 2023 14:21:45 GMT
• Title: Infinite Class Mixup
• Authors: Thomas Mensink, Pascal Mettes
• Abstract summary: Mixup is a strategy for training deep networks where additional samples are augmented by interpolating inputs and labels of training pairs. This paper seeks to address this cornerstone by mixing the classifiers directly instead of mixing the labels for each mixed pair. We show that Infinite Class Mixup outperforms standard Mixup and variants such as RegMixup and Remix on balanced, long-tailed, and data-constrained benchmarks.
• Score: 26.48101652432502
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Mixup is a widely adopted strategy for training deep networks, where additional samples are augmented by interpolating inputs and labels of training pairs. Mixup has shown to improve classification performance, network calibration, and out-of-distribution generalisation. While effective, a cornerstone of Mixup, namely that networks learn linear behaviour patterns between classes, is only indirectly enforced since the output interpolation is performed at the probability level. This paper seeks to address this limitation by mixing the classifiers directly instead of mixing the labels for each mixed pair. We propose to define the target of each augmented sample as a uniquely new classifier, whose parameters are a linear interpolation of the classifier vectors of the input pair. The space of all possible classifiers is continuous and spans all interpolations between classifier pairs. To make optimisation tractable, we propose a dual-contrastive Infinite Class Mixup loss, where we contrast the classifier of a mixed pair to both the classifiers and the predicted outputs of other mixed pairs in a batch. Infinite Class Mixup is generic in nature and applies to many variants of Mixup. Empirically, we show that it outperforms standard Mixup and variants such as RegMixup and Remix on balanced, long-tailed, and data-constrained benchmarks, highlighting its broad applicability.

Related papers

• Mixup Augmentation with Multiple Interpolations [26.46413903248954]
  We propose a simple yet effective extension called multi-mix, which generates multiple gradients from a sample pair. With an ordered sequence of generated samples, multi-mix can better guide the training process than standard mixup.
  arXiv  Detail & Related papers  (2024-06-03T15:16:09Z)
• Adversarial AutoMixup [50.1874436169571]
  We propose AdAutomixup, an adversarial automatic mixup augmentation approach. It generates challenging samples to train a robust classifier for image classification. Our approach outperforms the state of the art in various classification scenarios.
  arXiv  Detail & Related papers  (2023-12-19T08:55:00Z)
• Semantic Equivariant Mixup [54.734054770032934]
  Mixup is a well-established data augmentation technique, which can extend the training distribution and regularize the neural networks. Previous mixup variants tend to over-focus on the label-related information. We propose a semantic equivariant mixup (sem) to preserve richer semantic information in the input.
  arXiv  Detail & Related papers  (2023-08-12T03:05:53Z)
• The Implicit Bias of Batch Normalization in Linear Models and Two-layer Linear Convolutional Neural Networks [117.93273337740442]
  We show that gradient descent converges to a uniform margin classifier on the training data with an $exp(-Omega(log2 t))$ convergence rate. We also show that batch normalization has an implicit bias towards a patch-wise uniform margin.
  arXiv  Detail & Related papers  (2023-06-20T16:58:00Z)
• Selective Mixup Helps with Distribution Shifts, But Not (Only) because of Mixup [26.105340203096596]
  We show that non-random selection of pairs affects the training distribution and improve generalization by means completely unrelated to the mixing. We have found a new equivalence between two successful methods: selective mixup and resampling.
  arXiv  Detail & Related papers  (2023-05-26T10:56:22Z)
• MixupE: Understanding and Improving Mixup from Directional Derivative Perspective [86.06981860668424]
  We propose an improved version of Mixup, theoretically justified to deliver better generalization performance than the vanilla Mixup. Our results show that the proposed method improves Mixup across multiple datasets using a variety of architectures.
  arXiv  Detail & Related papers  (2022-12-27T07:03:52Z)
• C-Mixup: Improving Generalization in Regression [71.10418219781575]
  Mixup algorithm improves generalization by linearly interpolating a pair of examples and their corresponding labels. We propose C-Mixup, which adjusts the sampling probability based on the similarity of the labels. C-Mixup achieves 6.56%, 4.76%, 5.82% improvements in in-distribution generalization, task generalization, and out-of-distribution robustness, respectively.
  arXiv  Detail & Related papers  (2022-10-11T20:39:38Z)
• Harnessing Hard Mixed Samples with Decoupled Regularizer [69.98746081734441]
  Mixup is an efficient data augmentation approach that improves the generalization of neural networks by smoothing the decision boundary with mixed data. In this paper, we propose an efficient mixup objective function with a decoupled regularizer named Decoupled Mixup (DM) DM can adaptively utilize hard mixed samples to mine discriminative features without losing the original smoothness of mixup.
  arXiv  Detail & Related papers  (2022-03-21T07:12:18Z)
• AutoMix: Unveiling the Power of Mixup [34.623943038648164]
  We present a flexible, general Automatic Mixup framework which utilizes discriminative features to learn a sample mixing policy adaptively. We regard mixup as a pretext task and split it into two sub-problems: mixed samples generation and mixup classification. Experiments on six popular classification benchmarks show that AutoMix consistently outperforms other leading mixup methods.
  arXiv  Detail & Related papers  (2021-03-24T07:21:53Z)

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.