-
Stylist: Style-Driven Feature Ranking for Robust Novelty Detection
Authors:
Stefan Smeu,
Elena Burceanu,
Emanuela Haller,
Andrei Liviu Nicolicioiu
Abstract:
Novelty detection aims at finding samples that differ in some form from the distribution of seen samples. But not all changes are created equal. Data can suffer a multitude of distribution shifts, and we might want to detect only some types of relevant changes. Similar to works in out-of-distribution generalization, we propose to use the formalization of separating into semantic or content changes…
▽ More
Novelty detection aims at finding samples that differ in some form from the distribution of seen samples. But not all changes are created equal. Data can suffer a multitude of distribution shifts, and we might want to detect only some types of relevant changes. Similar to works in out-of-distribution generalization, we propose to use the formalization of separating into semantic or content changes, that are relevant to our task, and style changes, that are irrelevant. Within this formalization, we define the robust novelty detection as the task of finding semantic changes while being robust to style distributional shifts. Leveraging pretrained, large-scale model representations, we introduce Stylist, a novel method that focuses on drop** environment-biased features. First, we compute a per-feature score based on the feature distribution distances between environments. Next, we show that our selection manages to remove features responsible for spurious correlations and improve novelty detection performance. For evaluation, we adapt domain generalization datasets to our task and analyze the methods behaviors. We additionally built a large synthetic dataset where we have control over the spurious correlations degree. We prove that our selection mechanism improves novelty detection algorithms across multiple datasets, containing both stylistic and content shifts.
△ Less
Submitted 5 October, 2023;
originally announced October 2023.
-
Environment-biased Feature Ranking for Novelty Detection Robustness
Authors:
Stefan Smeu,
Elena Burceanu,
Emanuela Haller,
Andrei Liviu Nicolicioiu
Abstract:
We tackle the problem of robust novelty detection, where we aim to detect novelties in terms of semantic content while being invariant to changes in other, irrelevant factors. Specifically, we operate in a setup with multiple environments, where we determine the set of features that are associated more with the environments, rather than to the content relevant for the task. Thus, we propose a meth…
▽ More
We tackle the problem of robust novelty detection, where we aim to detect novelties in terms of semantic content while being invariant to changes in other, irrelevant factors. Specifically, we operate in a setup with multiple environments, where we determine the set of features that are associated more with the environments, rather than to the content relevant for the task. Thus, we propose a method that starts with a pretrained embedding and a multi-env setup and manages to rank the features based on their environment-focus. First, we compute a per-feature score based on the feature distribution variance between envs. Next, we show that by drop** the highly scored ones, we manage to remove spurious correlations and improve the overall performance by up to 6%, both in covariance and sub-population shift cases, both for a real and a synthetic benchmark, that we introduce for this task.
△ Less
Submitted 10 October, 2023; v1 submitted 21 September, 2023;
originally announced September 2023.
-
Env-Aware Anomaly Detection: Ignore Style Changes, Stay True to Content!
Authors:
Stefan Smeu,
Elena Burceanu,
Andrei Liviu Nicolicioiu,
Emanuela Haller
Abstract:
We introduce a formalization and benchmark for the unsupervised anomaly detection task in the distribution-shift scenario. Our work builds upon the iWildCam dataset, and, to the best of our knowledge, we are the first to propose such an approach for visual data. We empirically validate that environment-aware methods perform better in such cases when compared with the basic Empirical Risk Minimizat…
▽ More
We introduce a formalization and benchmark for the unsupervised anomaly detection task in the distribution-shift scenario. Our work builds upon the iWildCam dataset, and, to the best of our knowledge, we are the first to propose such an approach for visual data. We empirically validate that environment-aware methods perform better in such cases when compared with the basic Empirical Risk Minimization (ERM). We next propose an extension for generating positive samples for contrastive methods that considers the environment labels when training, improving the ERM baseline score by 8.7%.
△ Less
Submitted 23 November, 2022; v1 submitted 6 October, 2022;
originally announced October 2022.
