iNNvestigate neural networks!

Abstract

In recent years, deep neural networks have revolutionized many application domains of machine learning and are key components of many critical decision or predictive processes. Therefore, it is crucial that domain specialists can understand and analyze actions and pre- dictions, even of the most complex neural network architectures. Despite these arguments neural networks are often treated as black boxes. In the attempt to alleviate this short- coming many analysis methods were proposed, yet the lack of reference implementations often makes a systematic comparison between the methods a major effort. The presented library iNNvestigate addresses this by providing a common interface and out-of-the- box implementation for many analysis methods, including the reference implementation for PatternNet and PatternAttribution as well as for LRP-methods. To demonstrate the versatility of iNNvestigate, we provide an analysis of image classifications for variety of state-of-the-art neural network architectures.

References in corpus (6)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Deep Learning in Neural Networks: An Overview
• WaveNet: A Generative Model for Raw Audio
• Methods for Interpreting and Understanding Deep Neural Networks
• SmoothGrad: removing noise by adding noise
• Learning how to explain neural networks: PatternNet and PatternAttribution

Cited by in corpus (12)

• explAIner: A Visual Analytics Framework for Interactive and Explainable Machine Learning
• Local and Global Explanations of Agent Behavior: Integrating Strategy Summaries with Saliency Maps
• A Rate-Distortion Framework for Explaining Neural Network Decisions
• Modeling assembly bias with machine learning and symbolic regression
• Evolved Explainable Classifications for Lymph Node Metastases
• MeLIME: Meaningful Local Explanation for Machine Learning Models
• A Simple Saliency Method That Passes the Sanity Checks
• A psychophysics approach for quantitative comparison of interpretable computer vision models
• FIND: Human-in-the-Loop Debugging Deep Text Classifiers
• Human-grounded Evaluations of Explanation Methods for Text Classification
• Interpreting deep urban sound classification using Layer-wise Relevance Propagation
• Comparison of Convolutional neural network training parameters for detecting Alzheimers disease and effect on visualization