Collaborative Computing Optimization in Train-Edge-Cloud-Based Smart Train Systems Using Risk-Sensitive Reinforcement Learning
Bibliographic record
Abstract
With the advent of the intelligent and digital era, intelligent urban rail transit systems have been a research focus. As the core part of intelligent urban rail transit systems, smart trains are empowered by various intelligent applications. While improving system performance and reducing system risk, intelligent applications demand a large amount of computing power. However, it is challenging to provide simultaneously all intelligent applications for smart trains due to limited on-board computing resources. In this article, we design a train-edge-cloud (TEC) collaborative computing framework for train intelligent computing tasks. We aim to develop a TEC-based collaborative computing scheme to minimize the task processing delay with edge computing resource constraints. Considering the unique environment of smart train systems, we design a risk-sensitive reinforcement learning (RL) algorithm to realize collaborative computing optimization. We design a novel risk function in the system by jointly considering the computing load of edge intelligence (EI) servers and the characteristics of the urban rail transit systems. Moreover, we optimize the proposed risk-sensitive RL algorithm by using quantum representation and functions to accelerate its convergence speed. We design the TEC-based collaborative computing framework and design the quantum-inspired risk-sensitive RL algorithm to formulate the strategies for task scheduling. Comprehensive simulation results indicate that the algorithm adopted in this article can significantly reduce the task processing delay while satisfying EI servers' computing resource constraints. The quantum-inspired-optimized risk-sensitive RL model dramatically improves the model convergence speed.
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 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 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) | 0.002 | 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".