Balanced Activation for Long-tailed Visual Recognition

• URL: http://arxiv.org/abs/2008.11037v1
• Date: Mon, 24 Aug 2020 11:36:10 GMT
• Title: Balanced Activation for Long-tailed Visual Recognition
• Authors: Jiawei Ren, Cunjun Yu, Zhongang Cai, Haiyu Zhao
• Abstract summary: We introduce Balanced Activation to accommodate the label distribution shift between training and testing in object detection. We show that Balanced Activation generally provides 3% gain in terms of mAP on LVIS-1.0 and outperforms the current state-of-the-art methods without introducing any extra parameters.
• Score: 13.981652331491558
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep classifiers have achieved great success in visual recognition. However, real-world data is long-tailed by nature, leading to the mismatch between training and testing distributions. In this report, we introduce Balanced Activation (Balanced Softmax and Balanced Sigmoid), an elegant unbiased, and simple extension of Sigmoid and Softmax activation function, to accommodate the label distribution shift between training and testing in object detection. We derive the generalization bound for multiclass Softmax regression and show our loss minimizes the bound. In our experiments, we demonstrate that Balanced Activation generally provides ~3% gain in terms of mAP on LVIS-1.0 and outperforms the current state-of-the-art methods without introducing any extra parameters.

Related papers

• Strike a Balance in Continual Panoptic Segmentation [60.26892488010291]
  We introduce past-class backtrace distillation to balance the stability of existing knowledge with the adaptability to new information. We also introduce a class-proportional memory strategy, which aligns the class distribution in the replay sample set with that of the historical training data. We present a new method named Continual Panoptic Balanced (BalConpas)
  arXiv  Detail & Related papers  (2024-07-23T09:58:20Z)
• Uncertainty Aware Learning for Language Model Alignment [97.36361196793929]
  We propose uncertainty-aware learning (UAL) to improve the model alignment of different task scenarios. We implement UAL in a simple fashion -- adaptively setting the label smoothing value of training according to the uncertainty of individual samples. Experiments on widely used benchmarks demonstrate that our UAL significantly and consistently outperforms standard supervised fine-tuning.
  arXiv  Detail & Related papers  (2024-06-07T11:37:45Z)
• Rethinking Classifier Re-Training in Long-Tailed Recognition: A Simple Logits Retargeting Approach [102.0769560460338]
  We develop a simple logits approach (LORT) without the requirement of prior knowledge of the number of samples per class. Our method achieves state-of-the-art performance on various imbalanced datasets, including CIFAR100-LT, ImageNet-LT, and iNaturalist 2018.
  arXiv  Detail & Related papers  (2024-03-01T03:27:08Z)
• Bias Mitigating Few-Shot Class-Incremental Learning [17.185744533050116]
  Few-shot class-incremental learning aims at recognizing novel classes continually with limited novel class samples. Recent methods somewhat alleviate the accuracy imbalance between base and incremental classes by fine-tuning the feature extractor in the incremental sessions. We propose a novel method to mitigate model bias of the FSCIL problem during training and inference processes.
  arXiv  Detail & Related papers  (2024-02-01T10:37:41Z)
• 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)
• LiDAL: Inter-frame Uncertainty Based Active Learning for 3D LiDAR Semantic Segmentation [36.69417080183985]
  LiDAL is a novel active learning method for 3D LiDAR semantic segmentation. We propose LiDAL by exploiting inter-frame uncertainty among LiDAR frames. We achieve 95% of the performance of fully supervised learning with less than 5% of annotations.
  arXiv  Detail & Related papers  (2022-11-11T04:47:33Z)
• Regularizing Variational Autoencoder with Diversity and Uncertainty Awareness [61.827054365139645]
  Variational Autoencoder (VAE) approximates the posterior of latent variables based on amortized variational inference. We propose an alternative model, DU-VAE, for learning a more Diverse and less Uncertain latent space.
  arXiv  Detail & Related papers  (2021-10-24T07:58:13Z)
• Influential Rank: A New Perspective of Post-training for Robust Model against Noisy Labels [23.80449026013167]
  We propose a new approach for learning from noisy labels (LNL) via post-training. We exploit the overfitting property of a trained model to identify mislabeled samples. Our post-training approach creates great synergies when combined with the existing LNL methods.
  arXiv  Detail & Related papers  (2021-06-14T08:04:18Z)
• Balanced Meta-Softmax for Long-Tailed Visual Recognition [46.215759445665434]
  We show that the Softmax function, though used in most classification tasks, gives a biased gradient estimation under the long-tailed setup. This paper presents Balanced Softmax, an elegant unbiased extension of Softmax, to accommodate the label distribution shift between training and testing. In our experiments, we demonstrate that Balanced Meta-Softmax outperforms state-of-the-art long-tailed classification solutions on both visual recognition and instance segmentation tasks.
  arXiv  Detail & Related papers  (2020-07-21T12:05:00Z)

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.