arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.04081cs.LGcs.AI
收藏

权重空间网络通过增强实现泛化改进

Improved Generalization of Weight Space Networks via Augmentations

Bar-Ilan University(巴伊兰大学);University of Amsterdam(阿姆斯特丹大学);Samsung - SAIT AI Lab, Montreal(三星- SAIT AI实验室蒙特利尔分部);NVIDIA Research(英伟达研究院);Technion(以色列理工学院)

Aviv Shamsian, Aviv Navon, David W. Zhang, Yan Zhang, Ethan Fetaya, Gal Chechik, Haggai Maron

AI总结:

针对权重空间网络过拟合问题,提出权重空间数据增强与MixUp方法,在分类和自监督学习中显著提升泛化性能。

摘要:

深度权重空间(DWS)学习是一个新兴的研究方向,其中神经网络处理其他神经网络的权重,其应用包括二维和三维神经场(INRs、NeRFs),以及对其他类型神经网络进行推断。不幸的是,权重空间模型往往遭受严重的过拟合。我们实证分析了过拟合的原因,发现关键原因是DWS数据集缺乏多样性。虽然一个给定对象可以由许多不同的权重配置表示,但典型的INR训练集未能捕捉到表示同一对象的INR之间的变异性。为了解决这个问题,我们探索了权重空间中的数据增强策略,并提出了一种适用于权重空间的MixUp方法。我们在两种设置中证明了这些方法的有效性。在分类任务中,它们带来的性能提升相当于拥有多达10倍的数据。在自监督对比学习中,它们在下游分类任务中取得了5-10%的显著增益。

英文摘要:

Learning in deep weight spaces (DWS), where neural networks process the weights of other neural networks, is an emerging research direction, with applications to 2D and 3D neural fields (INRs, NeRFs), as well as making inferences about other types of neural networks. Unfortunately, weight space models tend to suffer from substantial overfitting. We empirically analyze the reasons for this overfitting and find that a key reason is the lack of diversity in DWS datasets. While a given object can be represented by many different weight configurations, typical INR training sets fail to capture variability across INRs that represent the same object. To address this, we explore strategies for data augmentation in weight spaces and propose a MixUp method adapted for weight spaces. We demonstrate the effectiveness of these methods in two setups. In classification, they improve performance similarly to having up to 10 times more data. In self-supervised contrastive learning, they yield substantial 5-10% gains in downstream classification.

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。

↑