Manitest: Are classifiers really invariant?
Open accessThe DOI matches its Crossref record; 7 of the 7 references it cites were found in a registry or an index.
Abstract
Invariance to geometric transformations is a highly desirable property of automatic classifiers in many image recognition tasks. Nevertheless, it is unclear to which extent state-of-the-art classifiers are invariant to basic transformations such as rotations and translations. This is mainly due to the lack of general methods that properly measure such an invariance. In this paper, we propose a rigorous and systematic approach for quantifying the invariance to geometric transformations of any classifier. Our key idea is to cast the problem of assessing a classifier's invariance as the computation of geodesics along the manifold of transformed images. We propose the Manitest method, built on the efficient Fast Marching algorithm to compute the invariance of classifiers. Our new method quantifies in particular the importance of data augmentation for learning invariance from data, and the increased invariance of convolutional neural networks with depth. We foresee that the proposed generic tool for measuring invariance to a large class of geometric transformations and arbitrary classifiers will have many applications for evaluating and comparing classifiers based on their invariance, and help improving the invariance of existing classifiers.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
A note on two problems in connexion with graphs
Dijkstra, E. W. (1959) · Numerische Mathematik · doi:10.1007/bf01386390 · Cited by 29,065
A Global Geometric Framework for Nonlinear Dimensionality Reduction
Tenenbaum, J. B., Silva, V. D., & Langford, J. C. (2000) · Science · doi:10.1126/science.290.5500.2319 · Cited by 13,585 · cited 1×
“This example underlines a well-known fundamental issue with the L 2 distance that fails to capture the intrinsic distance of the curved manifold of transformed images (see e.g., [9, 34]).”
BackgroundMatConvNet: Convolutional Neural Networks for MATLAB
Vedaldi, A., & Lenc, K. (2015) · Proceedings of the 23rd ACM international conference on Multimedia · doi:10.1145/2733373.2807412 · Cited by 2,444
Invariant Scattering Convolution Networks
Bruna, J., & Mallat, S. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2012.230 · Cited by 1,633
Learning methods for generic object recognition with invariance to pose and lighting
LeCun, Y., Huang, F. J., & Bottou, L. (2004) · Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. · doi:10.1109/cvpr.2004.1315150 · Cited by 1,324
Understanding image representations by measuring their equivariance and equivalence
Lenc, K., & Vedaldi, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298701 · Cited by 366
Cited by
DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks
Moosavi-Dezfooli, S.-M., Fawzi, A., & Frossard, P. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.282 · Cited by 5,317
“Similarly to the binary case, we first present the proposed approach for the linear case and then we generalize it to other classifiers.”
MethodSelf-citationDo ImageNet Classifiers Generalize to ImageNet?
Recht, B., Roelofs, R., Schmidt, L., & Shankar, V. (2019) · arXiv · doi:10.48550/arxiv.1902.10811 · Cited by 790
Why do deep convolutional networks generalize so poorly to small image transformations?
Azulay, A., & Weiss, Y. (2018) · arXiv · doi:10.48550/arxiv.1805.12177 · Cited by 428
“method to probe the invariances measures the linearity of the learned representations under natural transformations to the input image [28, 22, 16, 9].”
MethodRobustness May Be at Odds with Accuracy
Tsipras, D., Santurkar, S., Engstrom, L., Turner, A., & Mądry, A. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.12152 · Cited by 327
On the Robustness of Semantic Segmentation Models to Adversarial Attacks
Arnab, A., Miksik, O., & Torr, P. H. S. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00099 · Cited by 319
Analysis of classifiers’ robustness to adversarial perturbations
Fawzi, A., Fawzi, O., & Frossard, P. (2018) · Machine Learning · doi:10.1007/s10994-017-5663-3 · Cited by 287
Self-citationAdaptive data augmentation for image classification
Fawzi, A., Samulowitz, H., Turaga, D., & Frossard, P. (2016) · 2016 IEEE International Conference on Image Processing (ICIP) · doi:10.1109/icip.2016.7533048 · Cited by 239
“Index Terms— Data augmentation, transformation invariance, image robustness, trust-region optimization.”
BackgroundSelf-citationMeasuring Robustness to Natural Distribution Shifts in Image Classification
Taori, R., Dave, A., Shankar, V., Carlini, N., Recht, B., & Schmidt, L. (2020) · arXiv · doi:10.48550/arxiv.2007.00644 · Cited by 206
SparseFool: A Few Pixels Make a Big Difference
Modas, A., Moosavi-Dezfooli, S.-M., & Frossard, P. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00930 · Cited by 205
“Interestingly, the existence of such adversarial perturbations is not only limited to additive perturbations [21, 12, 48] or classification tasks [8], but can be found in many other applications [47, 7, 28, 31, 39, 41, 42].”
BackgroundSelf-citationDo Image Classifiers Generalize Across Time?
Shankar, V., Dave, A., Roelofs, R., Ramanan, D., Recht, B., & Schmidt, L. (2019) · arXiv · doi:10.48550/arxiv.1906.02168 · Cited by 37
“Motivated by the artificial nature of `p attacks, recent work has proposed more realistic image modifications such as small rotations & translations [2, 8, 9, 18], hue and color changes [16], and common image corruptions such as Gaussian blur and JPEG compression [11, 15].”
Background
Metadata
- Authors
- Fawzi, A., & Frossard, P.
- Venue
- Procedings of the British Machine Vision Conference 2015
- Publisher
- British Machine Vision Association
- Published
- 2015
- Pages
- 106.1-106.13
- OpenAlex
- W1580021837
- Cited by
- 81
- References
- 7
- FWCI
- 2.36
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.