Adjustable Multi-Objective Deep Reinforcement Learning-Based Edge User Allocation
Bibliographic record
Abstract
Multi-Access Edge Computing (MEC) is a popular and promising paradigm that allows service providers to serve their users from nearby servers. In order to fully leverage the advantages of MEC, the mapping between users and edge servers is of utmost importance for service providers. The Edge User Allocation (EUA) problem has been widely studied from the perspective of service providers with different objectives, e.g., maximizing the number of allocated users, respecting the latency threshold, minimizing overall system cost, etc. However, service providers tend to have dynamic priorities for different objectives over time. In certain situations, a service provider may opt to prioritize the minimization of their system cost at the expense of not meeting all their users expectations, or vice-versa, throughout a range of priority degrees. In this paper, we present a Deep Reinforcement Learning (DRL) approach for allocating users to edge servers according to dynamic priorities. We consider the online EUA problem where users arrive and depart dynamically, and propose a distributed solution that does not require full observation of all the servers to make allocation decisions. We offer a solution for both user-satisfaction and cost-effectiveness, while allowing the service provider to adjust their priority for each objective. A series of experiments have been conducted to evaluate the performance of our approach, under different priority degrees, against other baseline approaches. The results show the potential benefits of the proposed scheme in providing an adjustable multi-option solution for service providers.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".