DeCo-MeSC: Deep Compression-Based Memory-Constrained Split Computing Framework for Cooperative Inference of Neural Network
Bibliographic record
Abstract
Split computing (SC) of a deep neural network (DNN) across end nodes is a key enabling technology to realize energy-efficient and low-latency cooperative inference in wireless networks and Internet-of-Things (IoT). In this paper, we propose a novel SC framework based on the concept of deep compression (DC) of DNN considering the strictly limited memory footprint of a mobile device, namely,DeCo-MeSC. In our proposed DeCo-MeSC framework, an initial part of a target DNN (up to so-called the split layer) for the mobile device is compressed with the DC technique to satisfy the memory constraint, while the remaining part of the target DNN (after the split layer) for a cloud server is uncompressed, yet fine-tuned to compensate for the performance loss due to the compression of the initial part. Furthermore, the jointly optimal pair of the split layer and data rate is determined efficiently to maximize the end-to-end inference accuracy under both the end-to-end inference latency and memory constraints. Extensive experimental results demonstrate that the proposed scheme performs markedly better than the existing schemes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".