Causal Transportability for Visual Recognition
- URL: http://arxiv.org/abs/2204.12363v1
- Date: Tue, 26 Apr 2022 15:02:11 GMT
- Title: Causal Transportability for Visual Recognition
- Authors: Chengzhi Mao, Kevin Xia, James Wang, Hao Wang, Junfeng Yang, Elias
Bareinboim, Carl Vondrick
- Abstract summary: We show that standard classifiers fail because the association between images and labels is not transportable across settings.
We then show that the causal effect, which severs all sources of confounding, remains invariant across domains.
This motivates us to develop an algorithm to estimate the causal effect for image classification.
- Score: 70.13627281087325
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Visual representations underlie object recognition tasks, but they often
contain both robust and non-robust features. Our main observation is that image
classifiers may perform poorly on out-of-distribution samples because spurious
correlations between non-robust features and labels can be changed in a new
environment. By analyzing procedures for out-of-distribution generalization
with a causal graph, we show that standard classifiers fail because the
association between images and labels is not transportable across settings.
However, we then show that the causal effect, which severs all sources of
confounding, remains invariant across domains. This motivates us to develop an
algorithm to estimate the causal effect for image classification, which is
transportable (i.e., invariant) across source and target environments. Without
observing additional variables, we show that we can derive an estimand for the
causal effect under empirical assumptions using representations in deep models
as proxies. Theoretical analysis, empirical results, and visualizations show
that our approach captures causal invariances and improves overall
generalization.
Related papers
- Estimating Causal Effects from Learned Causal Networks [56.14597641617531]
We propose an alternative paradigm for answering causal-effect queries over discrete observable variables.
We learn the causal Bayesian network and its confounding latent variables directly from the observational data.
We show that this emphmodel completion learning approach can be more effective than estimand approaches.
arXiv Detail & Related papers (2024-08-26T08:39:09Z) - Identifiable Latent Neural Causal Models [82.14087963690561]
Causal representation learning seeks to uncover latent, high-level causal representations from low-level observed data.
We determine the types of distribution shifts that do contribute to the identifiability of causal representations.
We translate our findings into a practical algorithm, allowing for the acquisition of reliable latent causal representations.
arXiv Detail & Related papers (2024-03-23T04:13:55Z) - Fine-grained Recognition with Learnable Semantic Data Augmentation [68.48892326854494]
Fine-grained image recognition is a longstanding computer vision challenge.
We propose diversifying the training data at the feature-level to alleviate the discriminative region loss problem.
Our method significantly improves the generalization performance on several popular classification networks.
arXiv Detail & Related papers (2023-09-01T11:15:50Z) - Nonparametric Identifiability of Causal Representations from Unknown
Interventions [63.1354734978244]
We study causal representation learning, the task of inferring latent causal variables and their causal relations from mixtures of the variables.
Our goal is to identify both the ground truth latents and their causal graph up to a set of ambiguities which we show to be irresolvable from interventional data.
arXiv Detail & Related papers (2023-06-01T10:51:58Z) - Generalizable Information Theoretic Causal Representation [37.54158138447033]
We propose to learn causal representation from observational data by regularizing the learning procedure with mutual information measures according to our hypothetical causal graph.
The optimization involves a counterfactual loss, based on which we deduce a theoretical guarantee that the causality-inspired learning is with reduced sample complexity and better generalization ability.
arXiv Detail & Related papers (2022-02-17T00:38:35Z) - Clustering and Structural Robustness in Causal Diagrams [1.9766522384767222]
Clustering of variables is a natural way to reduce the size of the causal diagram.
We define a specific type of cluster, called transit cluster, that is guaranteed to preserve the identifiability properties of causal effects.
We show that this kind of structural robustness is closely related to transit clusters.
arXiv Detail & Related papers (2021-11-08T13:56:22Z) - Towards Robust Classification Model by Counterfactual and Invariant Data
Generation [7.488317734152585]
Spuriousness occurs when some features correlate with labels but are not causal.
We propose two data generation processes to reduce spuriousness.
Our data generations outperform state-of-the-art methods in accuracy when spurious correlations break.
arXiv Detail & Related papers (2021-06-02T12:48:29Z) - Nonlinear Invariant Risk Minimization: A Causal Approach [5.63479133344366]
We propose a learning paradigm that enables out-of-distribution generalization in the nonlinear setting.
We show identifiability of the data representation up to very simple transformations.
Extensive experiments on both synthetic and real-world datasets show that our approach significantly outperforms a variety of baseline methods.
arXiv Detail & Related papers (2021-02-24T15:38:41Z) - DecAug: Out-of-Distribution Generalization via Decomposed Feature
Representation and Semantic Augmentation [29.18840132995509]
deep learning often suffers from out-of-distribution (OoD) generalization.
We propose DecAug, a novel decomposed feature representation and semantic augmentation approach for OoD generalization.
We show that DecAug outperforms other state-of-the-art methods on various OoD datasets.
arXiv Detail & Related papers (2020-12-17T03:46:09Z) - Understanding Adversarial Examples from the Mutual Influence of Images
and Perturbations [83.60161052867534]
We analyze adversarial examples by disentangling the clean images and adversarial perturbations, and analyze their influence on each other.
Our results suggest a new perspective towards the relationship between images and universal perturbations.
We are the first to achieve the challenging task of a targeted universal attack without utilizing original training data.
arXiv Detail & Related papers (2020-07-13T05:00:09Z)
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.