A Performance Study of Deep Neural Network Representations of Interpretable ML on Edge Devices with AI Accelerators

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 18; p. 5681
• Main Authors: Schauer, Julian, Goodarzi, Payman, Morsch, Jannis, Schütze, Andreas
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 11.09.2025
• Subjects: AI accelerator; Algorithms; Classification; Data collection; Data processing; Data transmission; Edge computing; Energy consumption; energy efficiency; Feature selection; Field programmable gate arrays; interpretable ML; latency; Machine learning; Methods; Neural networks; Power; Python; Sensors; smart sensors; interpretable ML; edge computing; energy efficiency; latency; AI accelerator; smart sensors
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s25185681

With the rising adoption of machine learning (ML) and deep learning (DL) applications, the demand for deploying these algorithms closer to sensors has grown significantly, particularly in sensor-driven use cases such as predictive maintenance (PM) and condition monitoring (CM). This study investigated a novel application-oriented approach to representing interpretable ML inference as deep neural networks (DNNs) regarding the latency and energy efficiency on the edge, to tackle the problem of inefficient, high-effort, and uninterpretable-implementation ML algorithms. For this purpose, the interpretable deep neural network representation (IDNNRep) was integrated into an open-source interpretable ML toolbox to demonstrate the inference time and energy efficiency improvements. The goal of this work was to enable the utilization of generic artificial intelligence (AI) accelerators for interpretable ML algorithms to achieve efficient inference on edge hardware in smart sensor applications. This novel approach was applied to one regression and one classification task from the field of PM and validated by implementing the inference on the neural processing unit (NPU) of the QXSP-ML81 Single-Board Computer and the tensor processing unit (TPU) of the Google Coral. Different quantization levels of the implementation were tested against common Python and C++ implementations. The novel implementation reduced the inference time by up to 80% and the mean energy consumption by up to 76% at the lowest precision with only a 0.4% loss of accuracy compared to the C++ implementation. With the successful utilization of generic AI accelerators, the performance was further improved with a 94% reduction for both the inference time and the mean energy consumption.