Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ConceptLens: from Pixels to Understanding

Abhilekha Dalal, Pascal Hitzler

TL;DR

An overview of ConceptLens, its implementation, and its application in real-time visualization of neuron activations and error margins through bar charts is presented.

Abstract

ConceptLens is an innovative tool designed to illuminate the intricate workings of deep neural networks (DNNs) by visualizing hidden neuron activations. By integrating deep learning with symbolic methods, ConceptLens offers users a unique way to understand what triggers neuron activations and how they respond to various stimuli. The tool uses error-margin analysis to provide insights into the confidence levels of neuron activations, thereby enhancing the interpretability of DNNs. This paper presents an overview of ConceptLens, its implementation, and its application in real-time visualization of neuron activations and error margins through bar charts.

ConceptLens: from Pixels to Understanding

TL;DR

An overview of ConceptLens, its implementation, and its application in real-time visualization of neuron activations and error margins through bar charts is presented.

Abstract

ConceptLens is an innovative tool designed to illuminate the intricate workings of deep neural networks (DNNs) by visualizing hidden neuron activations. By integrating deep learning with symbolic methods, ConceptLens offers users a unique way to understand what triggers neuron activations and how they respond to various stimuli. The tool uses error-margin analysis to provide insights into the confidence levels of neuron activations, thereby enhancing the interpretability of DNNs. This paper presents an overview of ConceptLens, its implementation, and its application in real-time visualization of neuron activations and error margins through bar charts.
Paper Structure (8 sections, 1 figure)

This paper contains 8 sections, 1 figure.

Table of Contents

  1. Introduction
  2. Method
  3. System Overview
  4. Error-Margin Analysis
  5. User Interface
  6. Technical Details
  7. Demonstration
  8. Conclusion

Figures (1)

  • Figure 1: ConceptLens visualizes detected concepts in images, showing their error margins. In this street scene, "cross_walk" and "road" are confidently recognized with low error percentages, while uncertainty is shown for other labels like "automobile" and "central_reservation".