Publication | International Conference on Machine Learning 2021
Robust Representation Learning via Perceptual Similarity Metrics
This paper introduces a general framework for learning representations from images/objects that are invariant to bias in data for downstream classification tasks.
Download publicationAbstract
Robust Representation Learning via Perceptual Similarity Metrics
Saeid Asgari Taghanaki, Kristy Choi, Amir Khasahmadi, Anirudh Goyal
International Conference on Machine Learning 2021
A fundamental challenge in artificial intelligence is learning useful representations of data that yield good performance on a downstream task, without overfitting to spurious input features. Extracting such task-relevant predictive information is particularly difficult for real-world datasets. In this work, we propose Contrastive Input Morphing (CIM), a representation learning framework that learns input-space transformations of the data to mitigate the effect of irrelevant input features on downstream performance. Our method leverages a perceptual similarity metric via a triplet loss to ensure that the transformation preserves task-relevant information.Empirically, we demonstrate the efficacy of our approach on tasks which typically suffer from the presence of spurious correlations: classification with nuisance information, out-of-distribution generalization, and preservation of subgroup accuracies. We additionally show that CIM is complementary to other mutual information-based representation learning techniques, and demonstrate that it improves the performance of variational information bottleneck (VIB) when used together.
Related Resources
See what’s new.
2003
Flows on Surfaces of Arbitrary TopologyIn this paper we introduce a method to simulate fluid flows on smooth…
2014
An End-to-End Approach to Making Self-Folded 3D Surface Shapes by Uniform HeatingThis paper presents an end-to-end approach for creating 3D shapes by…
2014
Towards Visualization of Simulated Occupants and their Interactions with Buildings at Multiple Time ScalesWhile most building simulation tools model occupancy using simple…
2023
Engineering a bridge that designs and builds itselfThe final article in our three-part series explores the manufacturing…
Get in touch
Something pique your interest? Get in touch if you’d like to learn more about Autodesk Research, our projects, people, and potential collaboration opportunities.
Contact us