For Better or For Worse? Learning Minimum Variance Features With Label Augmentation

• URL: http://arxiv.org/abs/2402.06855v3
• Date: Thu, 13 Feb 2025 03:10:12 GMT
• Title: For Better or For Worse? Learning Minimum Variance Features With Label Augmentation
• Authors: Muthu Chidambaram, Rong Ge,
• Abstract summary: In this work, we analyze the role played by the label augmentation aspect of data augmentation methods. We first prove that linear models on binary classification data trained with label augmentation learn only the minimum variance features in the data. We then use our techniques to show that even for nonlinear models and general data distributions, the label smoothing and Mixup losses are lower bounded by a function of the model output variance.
• Score: 7.183341902583164
• License:
• Abstract: Data augmentation has been pivotal in successfully training deep learning models on classification tasks over the past decade. An important subclass of data augmentation techniques - which includes both label smoothing and Mixup - involves modifying not only the input data but also the input label during model training. In this work, we analyze the role played by the label augmentation aspect of such methods. We first prove that linear models on binary classification data trained with label augmentation learn only the minimum variance features in the data, while standard training (which includes weight decay) can learn higher variance features. We then use our techniques to show that even for nonlinear models and general data distributions, the label smoothing and Mixup losses are lower bounded by a function of the model output variance. Lastly, we demonstrate empirically that this aspect of label smoothing and Mixup can be a positive and a negative. On the one hand, we show that the strong performance of label smoothing and Mixup on image classification benchmarks is correlated with learning low variance hidden representations. On the other hand, we show that Mixup and label smoothing can be more susceptible to low variance spurious correlations in the training data.

Related papers

• Learning from Noisy Labels via Self-Taught On-the-Fly Meta Loss Rescaling [6.861041888341339]
  We propose unsupervised on-the-fly meta loss rescaling to reweight training samples. We are among the first to attempt on-the-fly training data reweighting on the challenging task of dialogue modeling. Our strategy is robust in the face of noisy and clean data, handles class imbalance, and prevents overfitting to noisy labels.
  arXiv  Detail & Related papers  (2024-12-17T14:37:50Z)
• Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised Learning [44.91863420044712]
  In open-world semi-supervised learning, a machine learning model is tasked with uncovering novel categories from unlabeled data. We introduce 1) the adaptive synchronizing marginal loss which imposes class-specific negative margins to alleviate the model bias towards seen classes, and 2) the pseudo-label contrastive clustering which exploits pseudo-labels predicted by the model to group unlabeled data from the same category together. Our method balances the learning pace between seen and novel classes, achieving a remarkable 3% average accuracy increase on the ImageNet dataset.
  arXiv  Detail & Related papers  (2023-09-21T09:44:39Z)
• Stubborn Lexical Bias in Data and Models [50.79738900885665]
  We use a new statistical method to examine whether spurious patterns in data appear in models trained on the data. We apply an optimization approach to *reweight* the training data, reducing thousands of spurious correlations. Surprisingly, though this method can successfully reduce lexical biases in the training data, we still find strong evidence of corresponding bias in the trained models.
  arXiv  Detail & Related papers  (2023-06-03T20:12:27Z)
• Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labels [61.97359362447732]
  Learning from noisy labels is an important and long-standing problem in machine learning for real applications. In this paper, we reformulate the label-noise problem from a generative-model perspective. Our model achieves new state-of-the-art (SOTA) results on all the standard real-world benchmark datasets.
  arXiv  Detail & Related papers  (2023-05-31T03:01:36Z)
• Self-Evolution Learning for Mixup: Enhance Data Augmentation on Few-Shot Text Classification Tasks [75.42002070547267]
  We propose a self evolution learning (SE) based mixup approach for data augmentation in text classification. We introduce a novel instance specific label smoothing approach, which linearly interpolates the model's output and one hot labels of the original samples to generate new soft for label mixing up.
  arXiv  Detail & Related papers  (2023-05-22T23:43:23Z)
• SoftMatch: Addressing the Quantity-Quality Trade-off in Semi-supervised Learning [101.86916775218403]
  This paper revisits the popular pseudo-labeling methods via a unified sample weighting formulation. We propose SoftMatch to overcome the trade-off by maintaining both high quantity and high quality of pseudo-labels during training. In experiments, SoftMatch shows substantial improvements across a wide variety of benchmarks, including image, text, and imbalanced classification.
  arXiv  Detail & Related papers  (2023-01-26T03:53:25Z)
• NorMatch: Matching Normalizing Flows with Discriminative Classifiers for Semi-Supervised Learning [8.749830466953584]
  Semi-Supervised Learning (SSL) aims to learn a model using a tiny labeled set and massive amounts of unlabeled data. In this work we introduce a new framework for SSL named NorMatch. We demonstrate, through numerical and visual results, that NorMatch achieves state-of-the-art performance on several datasets.
  arXiv  Detail & Related papers  (2022-11-17T15:39:18Z)
• Cross-Model Pseudo-Labeling for Semi-Supervised Action Recognition [98.25592165484737]
  We propose a more effective pseudo-labeling scheme, called Cross-Model Pseudo-Labeling (CMPL) CMPL achieves $17.6%$ and $25.1%$ Top-1 accuracy on Kinetics-400 and UCF-101 using only the RGB modality and $1%$ labeled data, respectively.
  arXiv  Detail & Related papers  (2021-12-17T18:59:41Z)
• X-model: Improving Data Efficiency in Deep Learning with A Minimax Model [78.55482897452417]
  We aim at improving data efficiency for both classification and regression setups in deep learning. To take the power of both worlds, we propose a novel X-model. X-model plays a minimax game between the feature extractor and task-specific heads.
  arXiv  Detail & Related papers  (2021-10-09T13:56:48Z)
• MixRL: Data Mixing Augmentation for Regression using Reinforcement Learning [2.1345682889327837]
  Existing techniques for data augmentation largely focus on classification tasks and do not readily apply to regression tasks. We show that mixing examples that either have a large data or label distance may have an increasingly-negative effect on model performance. We propose MixRL, a data augmentation meta learning framework for regression that learns for each example how many nearest neighbors it should be mixed with for the best model performance.
  arXiv  Detail & Related papers  (2021-06-07T07:01:39Z)

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.