arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1912.03330cs.CVcs.LG
收藏

ClusterFit:改进可视化表示的泛化

ClusterFit: Improving Generalization of Visual Representations

Facebook AI(脸书人工智能研究院)

Xueting Yan, Ishan Misra, Abhinav Gupta, Deepti Ghadiyaram, Dhruv Mahajan

摘要:

具有弱监督和自监督策略的预训练卷积神经网络在几个计算机视觉任务中变得越来越流行。然而,由于缺乏强区分信号,这些学习的表示可能过拟合预训练目标(例如,主题标签预测),并且不能很好地推广到下游任务。在这项工作中,我们提出了一种简单的策略-给定一个数据集,我们(a)使用k-means对从预训练网络中提取的特征进行聚类,(b)使用聚类分配作为伪标签,在这个数据集上从头开始重新训练一个新的网络。我们的经验表明,聚类有助于减少从提取的特征中提取的预训练任务特定信息,从而最大限度地减少过拟合。我们的方法可扩展到不同的预训练框架-弱监督和自我监督,模态-图像和视频,以及预训练任务-对象和动作分类。通过对11个不同词汇和粒度的不同目标数据集进行广泛的迁移学习实验,我们表明,与最先进的大规模(数百万/数十亿)弱监督图像和视频模型以及自监督图像模型相比,QuinterFit显着提高了表示质量。

英文摘要:

Pre-training convolutional neural networks with weakly-supervised and self-supervised strategies is becoming increasingly popular for several computer vision tasks. However, due to the lack of strong discriminative signals, these learned representations may overfit to the pre-training objective (e.g., hashtag prediction) and not generalize well to downstream tasks. In this work, we present a simple strategy - ClusterFit (CF) to improve the robustness of the visual representations learned during pre-training. Given a dataset, we (a) cluster its features extracted from a pre-trained network using k-means and (b) re-train a new network from scratch on this dataset using cluster assignments as pseudo-labels. We empirically show that clustering helps reduce the pre-training task-specific information from the extracted features thereby minimizing overfitting to the same. Our approach is extensible to different pre-training frameworks -- weak- and self-supervised, modalities -- images and videos, and pre-training tasks -- object and action classification. Through extensive transfer learning experiments on 11 different target datasets of varied vocabularies and granularities, we show that ClusterFit significantly improves the representation quality compared to the state-of-the-art large-scale (millions / billions) weakly-supervised image and video models and self-supervised image models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑