Explaining Deep Learning Representations by Tracing the Training Process

TL;DR

This paper introduces a new explanation method for deep neural networks that traces how intermediate representations evolve during training, identifying influential training examples and class contributions.

Contribution

It presents a general approach applicable to various architectures and training procedures, enabling detailed analysis of training dynamics and decision explanations.

Findings

Identifies influential training examples for model decisions

Provides visualization of training process and class contributions

Works with both single-instance and mini-batch training

Abstract

We propose a novel explanation method that explains the decisions of a deep neural network by investigating how the intermediate representations at each layer of the deep network were refined during the training process. This way we can a) find the most influential training examples during training and b) analyze which classes attributed most to the final representation. Our method is general: it can be wrapped around any iterative optimization procedure and covers a variety of neural network architectures, including feed-forward networks and convolutional neural networks. We first propose a method for stochastic training with single training instances, but continue to also derive a variant for the common mini-batch training. In experimental evaluations, we show that our method identifies highly representative training instances that can be used as an explanation. Additionally, we propose a visualization that provides explanations in the form of aggregated statistics over the whole training process.