Towards Explainable Deep Neural Networks (xDNN)

TL;DR

This paper introduces xDNN, an explainable, efficient deep learning model using prototypes that outperforms traditional methods in accuracy and training time, with clear interpretability and no need for extensive computational resources.

Contribution

The paper presents xDNN, a novel non-iterative, prototype-based deep learning architecture that is explainable, computationally efficient, and achieves state-of-the-art results on benchmark datasets.

Findings

xDNN outperforms existing methods in accuracy and training speed.

xDNN achieves a world record on the Caltech-256 dataset.

The approach is fully explainable and requires minimal computational resources.

Abstract

In this paper, we propose an elegant solution that is directly addressing the bottlenecks of the traditional deep learning approaches and offers a clearly explainable internal architecture that can outperform the existing methods, requires very little computational resources (no need for GPUs) and short training times (in the order of seconds). The proposed approach, xDNN is using prototypes. Prototypes are actual training data samples (images), which are local peaks of the empirical data distribution called typicality as well as of the data density. This generative model is identified in a closed form and equates to the pdf but is derived automatically and entirely from the training data with no user- or problem-specific thresholds, parameters or intervention. The proposed xDNN offers a new deep learning architecture that combines reasoning and learning in a synergy. It is non-iterative and non-parametric, which explains its efficiency in terms of time and computational resources. From the user perspective, the proposed approach is clearly understandable to human users. We tested it on some well-known benchmark data sets such as iRoads and Caltech-256. xDNN outperforms the other methods including deep learning in terms of accuracy, time to train and offers a clearly explainable classifier. In fact, the result on the very hard Caltech-256 problem (which has 257 classes) represents a world record.