Identifying Class Specific Filters with L1 Norm Frequency Histograms in Deep CNNs

12/14/2021
by   Akshay Badola, et al.
0

Interpretability of Deep Neural Networks has become a major area of exploration. Although these networks have achieved state of the art accuracy in many tasks, it is extremely difficult to interpret and explain their decisions. In this work we analyze the final and penultimate layers of Deep Convolutional Networks and provide an efficient method for identifying subsets of features that contribute most towards the network's decision for a class. We demonstrate that the number of such features per class is much lower in comparison to the dimension of the final layer and therefore the decision surface of Deep CNNs lies on a low dimensional manifold and is proportional to the network depth. Our methods allow to decompose the final layer into separate subspaces which is far more interpretable and has a lower computational cost as compared to the final layer of the full network.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/23/2023

Take 5: Interpretable Image Classification with a Handful of Features

Deep Neural Networks use thousands of mostly incomprehensible features t...
research
10/30/2017

The loss surface and expressivity of deep convolutional neural networks

We analyze the expressiveness and loss surface of practical deep convolu...
research
08/05/2023

Neural Collapse in the Intermediate Hidden Layers of Classification Neural Networks

Neural Collapse (NC) gives a precise description of the representations ...
research
01/29/2021

The Mind's Eye: Visualizing Class-Agnostic Features of CNNs

Visual interpretability of Convolutional Neural Networks (CNNs) has gain...
research
12/10/2020

On the emergence of tetrahedral symmetry in the final and penultimate layers of neural network classifiers

A recent numerical study observed that neural network classifiers enjoy ...
research
09/24/2019

Layerwise Relevance Visualization in Convolutional Text Graph Classifiers

Representations in the hidden layers of Deep Neural Networks (DNN) are o...
research
12/04/2022

Neural Fourier Filter Bank

We present a novel method to provide efficient and highly detailed recon...

Please sign up or login with your details

Forgot password? Click here to reset