Mobility-Aware Computation Offloading for AR Tasks Over Terahertz Wireless Networks: An Offline Reinforcement Learning Approach
Bibliographic record
Abstract
Augmented reality (AR) holds great promise within the Internet of vehicles (IoV), offering real-time, context-aware information to enhance user experiences. Since wireless AR services have stringent requirements for latency and energy efficiency, terahertz (THz) transmission and mobile edge computing (MEC) can be leveraged to address these issues. Besides, reinforcement learning (RL) is widely employed to design resource allocation and offloading decision schemes, further boosting the benefits of THz and MEC communication. However, conventional RL-based schemes necessitate online trial-and-error interactions and policy updates, which may result in unsafe actions and pose risks, e.g., vehicle collisions. Inspired by this, we propose a novel static data-driven AR tasks offloading framework for the IoV with THz communication. The proposed framework utilizes an offline RL algorithm to train the offloading decision agents without real-world interactions. Specifically, we formulate an optimization problem of maximizing user experience by jointly optimizing bandwidth and computation resource allocation. To reduce the complexity of solving the original problem, we decompose it into two subproblems: 1) THz bandwidth allocation and 2) MEC offloading decisions. Considering the unique characteristics of THz communication that users' available bandwidth varies with their distances to the base station (BS), we design a heuristic THz bandwidth allocation algorithm to achieve fair allocation. To address the typical out-of-distribution (OOD) action issue encountered in offline RL, we exploit the conservative Q-learning (CQL) to make the optimal MEC offloading decisions enhancing user experience. Simulation results show that the proposed scheme outperforms state-of-the-art algorithms in terms of algorithm stability and user feedback.
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 | 0.002 |
| 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 | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".