Quantifying Spuriousness of Biased Datasets Using Partial Information Decomposition

• URL: http://arxiv.org/abs/2407.00482v1
• Date: Sat, 29 Jun 2024 16:05:47 GMT
• Title: Quantifying Spuriousness of Biased Datasets Using Partial Information Decomposition
• Authors: Barproda Halder, Faisal Hamman, Pasan Dissanayake, Qiuyi Zhang, Ilia Sucholutsky, Sanghamitra Dutta,
• Abstract summary: Spurious patterns refer to a mathematical association between two or more variables in a dataset that are not causally related. This work presents the first information-theoretic formalization of spuriousness in a dataset (given a split of spurious and core features) using a mathematical framework called Partial Information Decomposition (PID) We disentangle the joint information content that the spurious and core features share about another target variable into distinct components, namely unique, redundant, and synergistic information.
• Score: 14.82261635235695
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Spurious patterns refer to a mathematical association between two or more variables in a dataset that are not causally related. However, this notion of spuriousness, which is usually introduced due to sampling biases in the dataset, has classically lacked a formal definition. To address this gap, this work presents the first information-theoretic formalization of spuriousness in a dataset (given a split of spurious and core features) using a mathematical framework called Partial Information Decomposition (PID). Specifically, we disentangle the joint information content that the spurious and core features share about another target variable (e.g., the prediction label) into distinct components, namely unique, redundant, and synergistic information. We propose the use of unique information, with roots in Blackwell Sufficiency, as a novel metric to formally quantify dataset spuriousness and derive its desirable properties. We empirically demonstrate how higher unique information in the spurious features in a dataset could lead a model into choosing the spurious features over the core features for inference, often having low worst-group-accuracy. We also propose a novel autoencoder-based estimator for computing unique information that is able to handle high-dimensional image data. Finally, we also show how this unique information in the spurious feature is reduced across several dataset-based spurious-pattern-mitigation techniques such as data reweighting and varying levels of background mixing, demonstrating a novel tradeoff between unique information (spuriousness) and worst-group-accuracy.

Related papers

• Partial Information Decomposition for Data Interpretability and Feature Selection [3.7414804164475983]
  Partial Information Decomposition of Features (PIDF) is a new paradigm for simultaneous data interpretability and feature selection. We extensively evaluate PIDF using both synthetic and real-world data, demonstrating its potential applications and effectiveness.
  arXiv  Detail & Related papers  (2024-05-29T15:54:03Z)
• Common-Sense Bias Discovery and Mitigation for Classification Tasks [16.8259488742528]
  We propose a framework to extract feature clusters in a dataset based on image descriptions. The analyzed features and correlations are human-interpretable, so we name the method Common-Sense Bias Discovery (CSBD) Experiments show that our method discovers novel biases on multiple classification tasks for two benchmark image datasets.
  arXiv  Detail & Related papers  (2024-01-24T03:56:07Z)
• Debiasing Multimodal Models via Causal Information Minimization [65.23982806840182]
  We study bias arising from confounders in a causal graph for multimodal data. Robust predictive features contain diverse information that helps a model generalize to out-of-distribution data. We use these features as confounder representations and use them via methods motivated by causal theory to remove bias from models.
  arXiv  Detail & Related papers  (2023-11-28T16:46:14Z)
• Hodge-Aware Contrastive Learning [101.56637264703058]
  Simplicial complexes prove effective in modeling data with multiway dependencies. We develop a contrastive self-supervised learning approach for processing simplicial data.
  arXiv  Detail & Related papers  (2023-09-14T00:40:07Z)
• Stubborn Lexical Bias in Data and Models [50.79738900885665]
  We use a new statistical method to examine whether spurious patterns in data appear in models trained on the data. We apply an optimization approach to *reweight* the training data, reducing thousands of spurious correlations. Surprisingly, though this method can successfully reduce lexical biases in the training data, we still find strong evidence of corresponding bias in the trained models.
  arXiv  Detail & Related papers  (2023-06-03T20:12:27Z)
• Gacs-Korner Common Information Variational Autoencoder [102.89011295243334]
  We propose a notion of common information that allows one to quantify and separate the information that is shared between two random variables. We demonstrate that our formulation allows us to learn semantically meaningful common and unique factors of variation even on high-dimensional data such as images and videos.
  arXiv  Detail & Related papers  (2022-05-24T17:47:26Z)
• Unsupervised Features Ranking via Coalitional Game Theory for Categorical Data [0.28675177318965034]
  Unsupervised feature selection aims to reduce the number of features. We show that the deriving features' selection outperforms competing methods in lowering the redundancy rate.
  arXiv  Detail & Related papers  (2022-05-17T14:17:36Z)
• Equivariance Allows Handling Multiple Nuisance Variables When Analyzing Pooled Neuroimaging Datasets [53.34152466646884]
  In this paper, we show how bringing recent results on equivariant representation learning instantiated on structured spaces together with simple use of classical results on causal inference provides an effective practical solution. We demonstrate how our model allows dealing with more than one nuisance variable under some assumptions and can enable analysis of pooled scientific datasets in scenarios that would otherwise entail removing a large portion of the samples.
  arXiv  Detail & Related papers  (2022-03-29T04:54:06Z)
• Generating Data to Mitigate Spurious Correlations in Natural Language Inference Datasets [27.562256973255728]
  Natural language processing models often exploit spurious correlations between task-independent features and labels in datasets to perform well only within the distributions they are trained on. We propose to tackle this problem by generating a debiased version of a dataset, which can then be used to train a debiased, off-the-shelf model. Our approach consists of 1) a method for training data generators to generate high-quality, label-consistent data samples; and 2) a filtering mechanism for removing data points that contribute to spurious correlations.
  arXiv  Detail & Related papers  (2022-03-24T09:08:05Z)
• Capturing patterns of variation unique to a specific dataset [68.8204255655161]
  We propose a tuning-free method that identifies low-dimensional representations of a target dataset relative to one or more comparison datasets. We show in several experiments that UCA with a single background dataset achieves similar results compared to cPCA with various tuning parameters.
  arXiv  Detail & Related papers  (2021-04-16T15:07:32Z)

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.