Split-Et-Impera: A Framework for the Design of Distributed Deep Learning Applications

Department of Computer Science, University of Verona, Italy
I-SPLIT abstract
The Split-Et-Impera framework: i) Determine the saliency-based split point candidates. The inputs are fed into a neural network to extract the saliency maps using the Grad-CAM algorithm at each layer. Then, we average over all the maps to generate the final Cumulative Saliency (CS) curve with the candidate split points. ii) Simulate each candidate by reproducing its computation and its transmission. iii) Use the simulation statistics to find the best-split point satisfying the application constraints on accuracy and latency.

Abstract

Many recent pattern recognition applications rely on complex distributed architectures in which sensing and computational nodes interact together through a communication network. Deep neural networks (DNNs) play an important role in this scenario, furnishing powerful decision mechanisms, at the price of a high computational effort. Consequently, powerful state-of-the-art DNNs are frequently split over various computational nodes, e.g., a first part stays on an embedded device and the rest on a server. Deciding where to split a DNN is a challenge in itself, making the design of deep learning applications even more complicated. Therefore, we propose Split-Et-Impera, a novel and practical framework that i) determines the set of the best-split points of a neural network based on deep network interpretability principles without performing a tedious try-andtest approach, ii) performs a communication-aware simulation for the rapid evaluation of different neural network rearrangements, and iii) suggests the best match between the quality of service requirements of the application and the performance in terms of accuracy and latency time.

BibTeX

@inproceedings{capogrosso2023split,
    title={Split-Et-Impera: A Framework for the Design of Distributed Deep Learning Applications},
    author={Capogrosso, Luigi and Cunico, Federico and Lora, Michele and Cristani, Marco and Fummi, Franco and Quaglia, Davide},
    booktitle={2023 26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)},
    pages={39--44},
    year={2023},
    organization={IEEE}
}