paper-with-me

홈 › Papers

Studying Invariances of Trained Convolutional Neural Networks

2018-03-15 · Charlotte Bunne, Lukas Rahmann, Thomas Wolf

Convolutional Neural Networks (CNNs) define an exceptionally powerful class of models for image classification, but the theoretical background and the understanding of how invariances to certain transformations are learned is limited. In a large scale screening with images modified by different affine and nonaffine transformations of varying magnitude, we analyzed the behavior of the CNN architectures AlexNet and ResNet. If the magnitude of different transformations does not exceed a class- and transformation dependent threshold, both architectures show invariant behavior. In this work we furthermore introduce a new learnable module, the Invariant Transformer Net, which enables us to learn differentiable parameters for a set of affine transformations. This allows us to extract the space of transformations to which the CNN is invariant and its class prediction robust.

📄 PDF Abstract BibTeX arXiv:1803.05963

Code (0)

등록된 구현이 없습니다.

Tasks

General Classificationimage-classificationImage Classification

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Average Pooling 설명 없음
Global Average Pooling Global Average Pooling is a pooling operation designed to replace fully connected layers in classical CNNs. The idea is to generate one feature map for each corresponding…
1x1 Convolution A 1 x 1 Convolution is a convolution with some special properties in that it can be used for dimensionality reduction,…
Batch Normalization 설명 없음
Bottleneck Residual Block A Bottleneck Residual Block is a variant of the residual block that utilises 1x1 convolutions to create a bottleneck. The…

Similar Papers 제목 키워드 기반

Parametrizing filters of a CNN with a GAN

2017-10-31 · ICLR 2018 1 · Yannic Kilcher, Gary Becigneul, Thomas Hofmann

It is commonly agreed that the use of relevant invariances as a good statistical bias is important in machine-learning. However, most approaches that explicitly incorporate invariances into a model architecture only make…

Generative Adversarial Network

Measuring Representational Robustness of Neural Networks Through Shared Invariances

2022-06-23 · Vedant Nanda, Till Speicher, Camila Kolling, John P. Dickerson 외

A major challenge in studying robustness in deep learning is defining the set of ``meaningless'' perturbations to which a given Neural Network (NN) should be invariant. Most work on robustness implicitly uses a human as …

Learning Invariances in Neural Networks

2020-10-22 · Gregory Benton, Marc Finzi, Pavel Izmailov, Andrew Gordon Wilson

Invariances to translations have imbued convolutional neural networks with powerful generalization properties. However, we often do not know a priori what invariances are present in the data, or to what extent a model sh…

image-classificationImage ClassificationMolecular Property PredictionProperty Prediction+1

Learning Invariances in Neural Networks from Training Data

2020-12-01 · NeurIPS 2020 12 · Gregory Benton, Marc Finzi, Pavel Izmailov, Andrew Gordon Wilson

Invariances to translations have imbued convolutional neural networks with powerful generalization properties. However, we often do not know a priori what invariances are present in the data, or to what extent a model sh…

image-classificationImage ClassificationMolecular Property PredictionProperty Prediction

Understanding Cross-Model Perceptual Invariances Through Ensemble Metamers

2025-04-02 · Lukas Boehm, Jonas Leo Mueller, Christoffer Loeffler, Leo Schwinn 외

Understanding the perceptual invariances of artificial neural networks is essential for improving explainability and aligning models with human vision. Metamers - stimuli that are physically distinct yet produce identica…