Split-Et-Impera: A Framework for the Design of Distributed Deep Learning Applications
This addresses the problem of designing distributed deep learning applications for developers and engineers by providing a practical framework, though it appears incremental as it builds on existing split computing methods.
The authors tackled the challenge of efficiently splitting deep neural networks across distributed computational nodes by proposing Split-Et-Impera, a framework that identifies optimal split points using interpretability principles, simulates network rearrangements, and matches application requirements to performance, resulting in improved design efficiency without specifying concrete numbers.
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-and-test 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.