Towards Explainable Deep Neural Networks (xDNN)

By Plamen Parvanov Angelov, Eduardo Almeida Soares. 2019

299 citations recorded by the source corpus.

In this paper, we propose an elegant solution that is directly addressing the bottlenecks of the traditional deep learning approaches and offers an 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 challenging problems as the classification of different lighting conditions for driving scenes (iROADS), object detection (Caltech-256, and Caltech-101), and SARS-CoV-2 identification via computed tomography scan (COVID CT-scans dataset). xDNN outperforms the other methods including deep learning in terms of accuracy, time to train and offers an explainable classifier.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Safety, Robustness & Interpretability

Citations in the corpus

Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges (2021, background): “Even for deep neural networks for computer vision, even on the most challenging of benchmark datasets, a plethora of machine learning techniques have been designed that do not sacrifice accuracy but gain substantial interpretability (Zhang et al., 2018; Chen et al., 2019; Koh et al., 2020; Chen et a”
A Survey on the Explainability of Supervised Machine Learning (2020, background): “Here, the parameters are often abstract and disconnected from the real world, which makes it difficult to interpret and explain the results of deep models (Angelov and Soares, 2019).”
Explainable artificial intelligence: an analytical review (2021, background): “…especially, DL) become more wide spread and intertwined with human-centric applications, and algorithmic decisions become more consequential to individuals and society, attention has shifted back from accuracy to explainability (Angelov & Soares, 2020; Core et al., 2006; Pedreschi et al., 2019).”
Towards Deep Machine Reasoning: a Prototype-based Deep Neural Network with Decision Tree Inference (2020, methodology): “In xDNN [1] we also followed this popular decision concept, but split it into two stages: i) per class winner, and ii) across classes global decision.”
Survey of Explainable AI Techniques in Healthcare (2023, background): “[38] ProtoPNet * * * [39] Triplet Network * * * * Example-based [5] xDNN * * *”
Trustworthy AI: From Principles to Practices (2021, methodology): “For example, previous research has embedded linear models and prototype selection to the DNN [16, 20, 73].”
Explainable AI (XAI): Core Ideas, Techniques, and Solutions (2022, background): “Inclusion of several layers in a neural network makes it difficult for designers to explain how the algorithm has reached a particular prediction outcome [1].”