Neural networks are known to use spurious correlations such as background information for classification. While prior work has looked at spurious correlations that are widespread in the training data, in this work, we investigate how sensitive neural networks are to rare spurious correlations, which may be harder to detect and correct, and may lead to privacy leaks. We introduce spurious patterns correlated with a fixed class to a few training examples and find that it takes only a handful of such examples for the network to learn the correlation. Furthermore, these rare spurious correlations also impact accuracy and privacy. We empirically and theoretically analyze different factors involved in rare spurious correlations and propose mitigation methods accordingly. Specifically, we observe that $\ell_2$ regularization and adding Gaussian noise to inputs can reduce the undesirable effects. Code available at https://github.com/yangarbiter/rare-spurious-correlation.
MM-SpuBench: Towards Better Understanding of Spurious Biases in Multimodal LLMs
University of Virginia(弗吉尼亚大学);Purdue University(普渡大学);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Georgia Institute of Technology(佐治亚理工学院)
Informativeness and Invariance: Two Perspectives on Spurious Correlations in Natural Language
Google Research(谷歌研究院)
伪相关性是对自然语言处理系统可信度的威胁,促使人们研究识别和消除伪相关性的方法。然而,解决虚假相关的问题需要更清楚地了解它们是什么以及它们如何在语言数据中出现。Gardner et al(2021)认为,语言的组成性质意味着标签和个体“输入特征”之间的\emph{all}相关性是虚假的。本文分析了这一建议的背景下,一个玩具的例子,展示了三个不同的条件,可以引起一个简单的PCFG的特征标签的相关性。将玩具示例链接到结构化因果模型表明:(1)即使标签对特征上的干预是不变的,特征-标签相关性也会出现;(2)即使标签对特征上的干预敏感,特征-标签相关性也可能不存在。由于输入特征将单独与标签相关,但在非常罕见的情况下,必须应用领域知识来识别虚假的相关性,构成真正的鲁棒性威胁。
Mila-Quebec AI Institute, Canada(魁北克AI研究所);Samsung - SAIT AI Lab, Montreal, Canada(三星SAIT人工智能实验室);University of Montreal, Canada(蒙特利尔大学);Institute of Science and Technology Austria(奥地利科学与技术研究所);National University of Singapore(新加坡国立大学)
Understanding and eliminating spurious modes in variational Monte Carlo using collective variables
Courant Institute of Mathematical Sciences, New York University(纽约大学库朗数学科学研究所);Division of Computing and Mathematical Sciences, California Institute of Technology(加州理工学院计算与数学科学系);Department of Mathematics, University of California, Berkeley(加州大学伯克利分校数学系);Department of Chemistry, Columbia University(哥伦比亚大学化学系);Center for Computational Quantum Physics, Flatiron Institute(熨斗研究所计算量子物理中心)
Understanding and Mitigating Spurious Signal Amplification in Test-Time Reinforcement Learning for Math Reasoning
NLPR & MAIS, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所NLPR与MAIS实验室);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);Meituan(美团);University of Science and Technology of China(中国科学技术大学)
Complex Dynamics in Simple Neural Networks: Understanding Gradient Flow in Phase Retrieval
Université Paris-Saclay(巴黎萨克雷大学);CNRS(法国国家科学研究中心);CEA(法国原子能和替代能源委员会);Institut de physique théorique(理论物理研究所);Laboratoire de Physique de l’Ecole normale supérieure ENS(巴黎高等师范学院物理实验室);Université PSL(巴黎文理研究大学);Sorbonne Université(索邦大学);Université Paris-Diderot(巴黎第七大学);Sorbonne Paris Cité(巴黎索邦西岱大学);King’s College London(伦敦国王学院)
Navigating Shortcuts, Spurious Correlations, and Confounders: From Origins via Detection to Mitigation
TU Darmstadt(达姆施塔特工业大学);Hessian Center for AI (hessian.AI)(黑森人工智能中心);German Research Center for AI (DFKI)(德国人工智能研究中心);Centre for Cognitive Science, TU Darmstadt(达姆施塔特工业大学认知科学中心)
不变风险最小化(Invariant Risk Minimization,ARJovsky et al.,2019)是最近提出的一个框架,旨在学习预测器,这些预测器在不同的训练环境中对虚假相关性是不变的。然而,尽管有理论上的理由,但在各种环境中,这种方法并没有得到广泛的测试。为了更好地理解该框架,我们使用IRMv1实证研究了几个研究问题,IRMv1是第一个提出来近似解决问题的实用算法。通过以不同的方式扩展ColoredMNIST实验,我们发现IRMv1(i)在训练环境之间的虚假相关性变化更大时表现更好,(ii)当底层关系近似不变时学习近似不变的预测器,(iii)可以扩展到文本分类的类似设置。
When PINNs Go Wrong: Pseudo-Time Stepping Against Spurious Solutions
Institute for Foundations of Data Science, Yale University(数据科学基础研究所,耶鲁大学);Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania(机械工程与应用力学系,宾夕法尼亚大学);Industrial Engineering & Management Science, Northwestern University(工业工程与管理科学,西北大学)
Robustar: Interactive Toolbox Supporting Precise Data Annotation for Robust Vision Learning
Carnegie Mellon University(卡内基梅隆大学);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);University of Nottingham Ningbo China(宁波诺丁汉大学);University of Nottingham(诺丁汉大学);University of California, Irvine(加州大学欧文分校);University of Michigan(密歇根大学);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)
Towards Robust Trajectory Representations: Isolating Environmental Confounders with Causal Learning
Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));University of Science and Technology of China(中国科学技术大学);Beijing Institute of Technology(北京理工大学)
Mitigating Shortcuts in Language Models with Soft Label Encoding
Stevens Institute of Technology(史蒂文斯理工学院);Shanghai Jiao Tong University(上海交通大学);New Jersey Institute of Technology(新泽西理工学院);University of Georgia(佐治亚大学)