Conference paper · 2026 · Open access

Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing

Kungui Zeng, Y Tian, Xiushuang Yi, Tao Xiong, Qiang He, Rui Wang

International Conference on Computer Supported Cooperative Work in Design (CSCWD), 2188–2193 · Published 13 May 2026 · DOI 10.1109/cscwd68734.2026.11582497

Open full textCite this paper

In brief

“Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing” is a 2026 conference paper by Kungui Zeng and 5 co-authors, published in International Conference on Computer Supported Cooperative Work in Design (CSCWD). OpenAlex counts 0 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at International Conference on Computer Supported Cooperative Work in Design (CSCWD).

Citations
0
References
13
Authors
6

Abstract

Mobile edge computing (MEC) alleviates the resource limitations of mobile devices by offloading computation-intensive tasks to nearby edge servers, while real-world applications are often modeled as workflows composed of interdependent subtasks. Most existing studies, however, overlook the dynamic temporal characteristics of task execution and lack effective guidance along critical paths in the task graph topology, making it difficult to anticipate global bottlenecks caused by resource contention and often leading to suboptimal completion time.To address these issues, this paper proposes a heuristic-guided spatio-temporal hybrid deep reinforcement learning method that aims to minimize the average task completion time under complex topological dependencies and dynamic resource constraints. Application workflows are modeled as directed acyclic graphs (DAGs) and the dependent task offloading problem is formulated as a Markov decision process (MDP). A spatio-temporal hybrid neural network, composed of a graph attention network (GAT) and a long short-term memory (LSTM) network, is employed to jointly capture high-dimensional topological features and temporal dependencies of the task graph. Furthermore, to cope with the non-stationarity of offloading decisions, a heuristic algorithm is incorporated as prior guidance and a collaborative training mechanism is developed to enhance policy learning. Simulation results demonstrate that, compared with baseline algorithms, the proposed method significantly reduces the average task completion time.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Zeng, K., Tian, Y., Yi, X., Xiong, T., He, Q., & Wang, R. (2026). Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing. International Conference on Computer Supported Cooperative Work in Design (CSCWD), 2188–2193. https://doi.org/10.1109/cscwd68734.2026.11582497

MLA 9

Zeng, Kungui, et al. “Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing.” International Conference on Computer Supported Cooperative Work in Design (CSCWD), 2026, pp. 2188–2193. https://doi.org/10.1109/cscwd68734.2026.11582497.

Chicago

Zeng, Kungui, Y Tian, Xiushuang Yi, Tao Xiong, Qiang He, and Rui Wang. 2026. “Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing.” International Conference on Computer Supported Cooperative Work in Design (CSCWD): 2188–2193. https://doi.org/10.1109/cscwd68734.2026.11582497.

Harvard

Zeng, K. et al. (2026) ‘Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing’, International Conference on Computer Supported Cooperative Work in Design (CSCWD), pp. 2188–2193. doi:10.1109/cscwd68734.2026.11582497.

IEEE

K. Zeng, Y. Tian, X. Yi, T. Xiong, Q. He, and R. Wang, “Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing,” International Conference on Computer Supported Cooperative Work in Design (CSCWD), pp. 2188–2193, 2026, doi: 10.1109/cscwd68734.2026.11582497.

BibTeX

@inproceedings{zeng2026spatio,
  title     = {{Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing}},
  author    = {Zeng, Kungui and Tian, Y and Yi, Xiushuang and Xiong, Tao and He, Qiang and Wang, Rui},
  booktitle = {International Conference on Computer Supported Cooperative Work in Design (CSCWD)},
  year      = {2026},
  pages     = {2188--2193},
  doi       = {10.1109/cscwd68734.2026.11582497},
  url       = {https://doi.org/10.1109/cscwd68734.2026.11582497}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing

  • mobile edge computing
  • task offloading
  • directed acyclic graphs
  • graph attention networks
  • long short-term memory
  • Markov decision process
  • heuristic guidance
  • collaborative training

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Zeng, K., Tian, Y., Yi, X., Xiong, T., He, Q., & Wang, R. (2026). Spatio-Temporal Hybrid Deep Reinforcement Learning for Dependent Task Offloading Optimization in Mobile Edge Computing. International Conference on Computer Supported Cooperative Work in Design (CSCWD), 2188–2193. https://doi.org/10.1109/cscwd68734.2026.11582497

How many times has this paper been cited?

0 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at International Conference on Computer Supported Cooperative Work in Design (CSCWD): sign in and open the full text from this page.

Who wrote this paper?

Kungui Zeng, Y Tian, Xiushuang Yi, Tao Xiong, Qiang He, Rui Wang.

What is the DOI of this paper?

10.1109/cscwd68734.2026.11582497, permanently at https://doi.org/10.1109/cscwd68734.2026.11582497. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.