Visualizing and Understanding Deep Neural Networks in CTR Prediction

TL;DR

This paper introduces visualization and analysis techniques for deep neural networks specifically applied to CTR prediction, aiding interpretability and model refinement in online advertising.

Contribution

It presents novel methods for visualizing and understanding deep neural networks in CTR prediction, including neuron-level inspection, layer-wise performance measurement, and feature influence analysis.

Findings

Neuron-level insights into model behavior

Layer-wise performance metrics

Feature saliency scores for input influence

Abstract

Although deep learning techniques have been successfully applied to many tasks, interpreting deep neural network models is still a big challenge to us. Recently, many works have been done on visualizing and analyzing the mechanism of deep neural networks in the areas of image processing and natural language processing. In this paper, we present our approaches to visualize and understand deep neural networks for a very important commercial task--CTR (Click-through rate) prediction. We conduct experiments on the productive data from our online advertising system with daily varying distribution. To understand the mechanism and the performance of the model, we inspect the model's inner status at neuron level. Also, a probe approach is implemented to measure the layer-wise performance of the model. Moreover, to measure the influence from the input features, we calculate saliency scores based on the back-propagated gradients. Practical applications are also discussed, for example, in understanding, monitoring, diagnosing and refining models and algorithms.