Identification of Systematic Errors of Image Classifiers on Rare Subgroups

• URL: http://arxiv.org/abs/2303.05072v2
• Date: Wed, 12 Apr 2023 10:05:41 GMT
• Title: Identification of Systematic Errors of Image Classifiers on Rare Subgroups
• Authors: Jan Hendrik Metzen, Robin Hutmacher, N. Grace Hua, Valentyn Boreiko, Dan Zhang
• Abstract summary: systematic errors can impact both fairness for demographic minority groups as well as robustness and safety under domain shift. We leverage recent advances in text-to-image models and search in the space of textual descriptions of subgroups ("prompts") for subgroups where the target model has low performance. We study subgroup coverage and identifiability with PromptAttack in a controlled setting and find that it identifies systematic errors with high accuracy.
• Score: 12.064692111429494
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Despite excellent average-case performance of many image classifiers, their performance can substantially deteriorate on semantically coherent subgroups of the data that were under-represented in the training data. These systematic errors can impact both fairness for demographic minority groups as well as robustness and safety under domain shift. A major challenge is to identify such subgroups with subpar performance when the subgroups are not annotated and their occurrence is very rare. We leverage recent advances in text-to-image models and search in the space of textual descriptions of subgroups ("prompts") for subgroups where the target model has low performance on the prompt-conditioned synthesized data. To tackle the exponentially growing number of subgroups, we employ combinatorial testing. We denote this procedure as PromptAttack as it can be interpreted as an adversarial attack in a prompt space. We study subgroup coverage and identifiability with PromptAttack in a controlled setting and find that it identifies systematic errors with high accuracy. Thereupon, we apply PromptAttack to ImageNet classifiers and identify novel systematic errors on rare subgroups.

Related papers

• Discover and Mitigate Multiple Biased Subgroups in Image Classifiers [45.96784278814168]
  Machine learning models can perform well on in-distribution data but often fail on biased subgroups that are underrepresented in the training data. We propose Decomposition, Interpretation, and Mitigation (DIM) to address this problem. Our approach decomposes the image features into multiple components that represent multiple subgroups.
  arXiv  Detail & Related papers  (2024-03-19T14:44:54Z)
• Uncovering Prototypical Knowledge for Weakly Open-Vocabulary Semantic Segmentation [59.37587762543934]
  This paper studies the problem of weakly open-vocabulary semantic segmentation (WOVSS) Existing methods suffer from a granularity inconsistency regarding the usage of group tokens. We propose the prototypical guidance network (PGSeg) that incorporates multi-modal regularization.
  arXiv  Detail & Related papers  (2023-10-29T13:18:00Z)
• Deep Hypothesis Tests Detect Clinically Relevant Subgroup Shifts in Medical Images [21.01688837312175]
  We focus on the detection of subgroup shifts in machine learning systems. Recent state-of-the-art statistical tests can be effectively applied to subgroup shift detection on medical imaging data.
  arXiv  Detail & Related papers  (2023-03-08T19:58:41Z)
• Outlier-Robust Group Inference via Gradient Space Clustering [50.87474101594732]
  Existing methods can improve the worst-group performance, but they require group annotations, which are often expensive and sometimes infeasible to obtain. We address the problem of learning group annotations in the presence of outliers by clustering the data in the space of gradients of the model parameters. We show that data in the gradient space has a simpler structure while preserving information about minority groups and outliers, making it suitable for standard clustering methods like DBSCAN.
  arXiv  Detail & Related papers  (2022-10-13T06:04:43Z)
• Subgroup Discovery in Unstructured Data [7.6323763630645285]
  Subgroup discovery has numerous applications in knowledge discovery and hypothesis generation. Subgroup-aware variational autoencoder learns a representation of unstructured data which leads to subgroups with higher quality.
  arXiv  Detail & Related papers  (2022-07-15T23:13:54Z)
• Addressing Missing Sources with Adversarial Support-Matching [8.53946780558779]
  We investigate a scenario in which the absence of certain data is linked to the second level of a two-level hierarchy in the data. Inspired by the idea of protected groups from algorithmic fairness, we refer to the partitions carved by this second level as "subgroups" We make use of an additional, diverse but unlabeled dataset, called the "deployment set", to learn a representation that is invariant to subgroup.
  arXiv  Detail & Related papers  (2022-03-24T16:19:19Z)
• Causal Scene BERT: Improving object detection by searching for challenging groups of data [125.40669814080047]
  Computer vision applications rely on learning-based perception modules parameterized with neural networks for tasks like object detection. These modules frequently have low expected error overall but high error on atypical groups of data due to biases inherent in the training process. Our main contribution is a pseudo-automatic method to discover such groups in foresight by performing causal interventions on simulated scenes.
  arXiv  Detail & Related papers  (2022-02-08T05:14:16Z)
• Towards Group Robustness in the presence of Partial Group Labels [61.33713547766866]
  spurious correlations between input samples and the target labels wrongly direct the neural network predictions. We propose an algorithm that optimize for the worst-off group assignments from a constraint set. We show improvements in the minority group's performance while preserving overall aggregate accuracy across groups.
  arXiv  Detail & Related papers  (2022-01-10T22:04:48Z)
• Learning Multi-Attention Context Graph for Group-Based Re-Identification [214.84551361855443]
  Learning to re-identify or retrieve a group of people across non-overlapped camera systems has important applications in video surveillance. In this work, we consider employing context information for identifying groups of people, i.e., group re-id. We propose a novel unified framework based on graph neural networks to simultaneously address the group-based re-id tasks.
  arXiv  Detail & Related papers  (2021-04-29T09:57:47Z)

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.