arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.08760cs.CVcs.AIcs.LG
收藏

解决自监督表示学习的低效性

Solving Inefficiency of Self-supervised Representation Learning

Sun Yat-sen University(中山大学);University of Oxford(牛津大学);DarkMatter AI Research(暗物质人工智能研究院);Nanyang Technological University(南洋理工大学)

Guangrun Wang, Keze Wang, Guangcong Wang, Philip H. S. Torr, Liang Lin

摘要:

自监督学习(特别是对比学习)由于其在无监督学习中的巨大潜力而引起了人们极大的兴趣。尽管取得了公认的成功,但现有的对比学习方法的学习效率非常低,例如,对于可比较的识别准确度,花费比监督学习多十倍的训练时期。本文揭示了对比学习中存在的两个矛盾现象,即欠聚类和过聚类问题,它们是影响学习效率的主要障碍。欠聚类意味着当用于对比学习的负样本对不足以区分所有实际对象类时,模型不能有效地学习以发现类间样本之间的相异性。过度聚类意味着模型不能有效地从过多的负样本对中学习特征,迫使模型将相同实际类别的样本过度聚类到不同的聚类中。为了同时克服这两个问题,我们提出了一种新的自监督学习框架,使用截断三重损失。准确地说,我们采用三重损失倾向于最大化正对和负对之间的相对距离,以解决欠聚类问题;并且我们通过从所有负样本中选择负样本代理来构建负对,以避免由伯努利分布模型保证的过聚类问题。我们在几个大规模的基准测试中广泛评估了我们的框架(例如,ImageNet、SYSU-30k和COCO)。结果证明了我们模型的优越性(例如,学习效率)明显优于最新的最先进的方法。代码可在https://github.com/wanggrun/triplet上获得。

英文摘要:

Self-supervised learning (especially contrastive learning) has attracted great interest due to its huge potential in learning discriminative representations in an unsupervised manner. Despite the acknowledged successes, existing contrastive learning methods suffer from very low learning efficiency, e.g., taking about ten times more training epochs than supervised learning for comparable recognition accuracy. In this paper, we reveal two contradictory phenomena in contrastive learning that we call under-clustering and over-clustering problems, which are major obstacles to learning efficiency. Under-clustering means that the model cannot efficiently learn to discover the dissimilarity between inter-class samples when the negative sample pairs for contrastive learning are insufficient to differentiate all the actual object classes. Over-clustering implies that the model cannot efficiently learn features from excessive negative sample pairs, forcing the model to over-cluster samples of the same actual classes into different clusters. To simultaneously overcome these two problems, we propose a novel self-supervised learning framework using a truncated triplet loss. Precisely, we employ a triplet loss tending to maximize the relative distance between the positive pair and negative pairs to address the under-clustering problem; and we construct the negative pair by selecting a negative sample deputy from all negative samples to avoid the over-clustering problem, guaranteed by the Bernoulli Distribution model. We extensively evaluate our framework in several large-scale benchmarks (e.g., ImageNet, SYSU-30k, and COCO). The results demonstrate our model's superiority (e.g., the learning efficiency) over the latest state-of-the-art methods by a clear margin. Codes available at: https://github.com/wanggrun/triplet .

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。

↑