Adaptive Multi-Task Learning for Urban Air Quality Assessment

Iman Youssif Ibrahim; Dindar M. Ahmed

doi:10.65542/djei.v2i3.83

Authors

Iman Youssif Ibrahim Department of Information Technology, Technical College of Informatics, Akre University for Applied Science, Duhok, Kurdistan Region, Iraq https://orcid.org/0009-0008-0637-7710
Dindar M. Ahmed Department of Information Technology, Technical College of Duhok, Duhok Polytechnic University, Duhok 42001, Kurdistan Region of Iraq https://orcid.org/0000-0002-1321-0221

DOI:

https://doi.org/10.65542/djei.v2i3.83

Keywords:

Multi-Task Learning, adaptive loss balancing, air quality assessment, PM2.5, PM10, AQI, tabular data modeling

Abstract

Accurate air pollution forecasting is vital for protecting the environment and public health. However, predicting air pollution continues to be a challenge due to complicated relationships between meteorological variables, pollutant concentrations, and time-dependent characteristics. This paper proposes a Temporal-Aware Multi-Task Learning (TMTL-AQI) framework to assess urban air quality via structured data. The model simultaneously performs three tasks: Air Quality Index category classification, PM2.5 and PM10 particulate matter regression, and auxiliary AQI value prediction. The framework consists of data preprocessing, cyclical temporal encoding, a shared backbone based on hard parameter sharing, and three task-specific output heads optimized jointly. It uses cyclical transformations to encode temporal traits that capture periodic patterns of the environment and produces a common neural representation through hard parameter sharing. The TRAQID dataset was used for empirical testing. The model outperforms single-task and baseline multi-task models with an accuracy of 0.7605 and an F1-score of 0.7422 for AQI classification. Although based solely on structured input (no image-based features), the model performed competitively with a state-of-the-art approach due to the reduced prediction error in regression tasks (MAE values of 13.34 μg/m³ for PM2.5 and 22.06 μg/m³ for PM10).

References

1. Health Effects Institute. State of Global Air 2025: A Report on Air Pollution and Its Role in the World’s Leading Causes of Death. Available online: https://www.stateofglobalair.org/sites/default/files/documents/2025-10/soga-2025-report.pdf (accessed on 17 August 2026).

2. European Environment Agency. Europe’s Air Quality Status 2023. Available online: https://www.eea.europa.eu/en/analysis/publications/europes-air-quality-status-2023 (accessed on 17 August 2026).

3. U.S. Environmental Protection Agency. Technical Assistance Document for the Reporting of Daily Air Quality—The Air Quality Index (AQI). Available online: https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=P101AP0Q.txt (accessed on 17 August 2026).

4. Zhang, B.; Rong, Y.; Yong, R.; Qin, D.; Li, M.; Zou, G.; Pan, J. Deep learning for air pollutant concentration prediction: A review. Atmospheric Environment 2022, 290, 119347. DOI: https://doi.org/10.1016/j.atmosenv.2022.119347

5. Jin, X.-B.; Wang, Z.-Y.; Kong, J.-L.; Bai, Y.-T.; Su, T.-L.; Ma, H.-J.; Chakrabarti, P. Deep spatio-temporal graph network with self-optimization for air quality prediction. Entropy 2023, 25, 247. DOI: https://doi.org/10.3390/e25020247

6. Chen, H.; Deng, G.; Liu, Y. Monitoring the influence of industrialization and urbanization on spatiotemporal variations of AQI and PM2.5 in three provinces, China. Atmosphere 2022, 13, 1377. DOI: https://doi.org/10.3390/atmos13091377

7. Wen, W.; Shen, L.; Sheng, L.; Ma, X.; Wang, J.; Guan, C.; Deng, G.; Li, H.; Zhou, B. Impact of meteorological uncertainties on PM2.5 forecast: An ensemble air quality forecast study during 2022 Beijing Winter Olympics. Atmospheric Environment 2025, 121000. DOI: https://doi.org/10.1016/j.atmosenv.2025.121027

8. Chang, Y.-S.; Chiao, H.-T.; Abimannan, S.; Huang, Y.-P.; Tsai, Y.-T.; Lin, K.-M. An LSTM-based aggregated model for air pollution forecasting. Atmospheric Pollution Research 2020, 11, 1451–1463. DOI: https://doi.org/10.1016/j.apr.2020.05.015

9. Liu, H.; Han, Q.; Sun, H.; Sheng, J.; Yang, Z. Spatiotemporal adaptive attention graph convolution network for city-level air quality prediction. Scientific Reports 2023, 13, 13462. DOI: https://doi.org/10.1038/s41598-023-39286-0

10. Xia, H.; Chen, X.; Wang, Z.; Chen, X.; Dong, F. A multi-modal deep-learning air quality prediction method based on multi-station time-series data and remote-sensing images: Case study of Beijing and Tianjin. Entropy 2024, 26, 91. DOI: https://doi.org/10.3390/e26010091

11. Xu, S.; Li, W.; Zhu, Y.; Xu, A. A novel hybrid model for six main pollutant concentrations forecasting based on improved LSTM neural networks. Scientific Reports 2022, 12, 14434. DOI: https://doi.org/10.1038/s41598-022-17754-3

12. Kathalkar, O.R.; Nilesh, N.N.; Chaudhari, S.; Namboodiri, A. TRAQID—Traffic-Related Air Quality Image Dataset. In Proceedings of the 15th Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP), Bengaluru, India, 13–15 December 2024. DOI: https://doi.org/10.1145/3702250.3702260

13. Abdelsamie, M.M.; Azab, S.S.; Hefny, H.A. Deep multi-task learning: A review of concepts, methods, and cross-domain applications. International Journal of Data Science and Analytics 2026, 21, 77. DOI: https://doi.org/10.1007/s41060-025-00892-y

14. Chen, X.; Zhang, Y.; Cao, K.; Li, D.; Wu, Q. Improving PM2.5 simulations using LSTM: A study on spatiotemporal generalization. Atmospheric Pollution Research 2025, 16, 102647. DOI: https://doi.org/10.1016/j.apr.2025.102647

15. Fang, Y.; Liu, S.; Su, Z. Air quality index prediction using a hybrid CEEMDAN-CNN-IGWO-BiGRU-Attention model. Scientific Reports 2026, 16, 15908. DOI: https://doi.org/10.1038/s41598-026-46978-w

16. Senushkin, D.; Patakin, N.; Kuznetsov, A.; Konushin, A. Independent Component Alignment for Multi-Task Learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, Canada, 18–22 June 2023. DOI: https://doi.org/10.1109/CVPR52729.2023.01923

17. Hervella, Á.S.; Rouco, J.; Novo, J.; Ortega, M. Multi-Adaptive Optimization for Multi-Task Learning with Deep Neural Networks. Neural Networks 2024, 170, 254–265. DOI: https://doi.org/10.1016/j.neunet.2023.11.038

18. Lakkapragada, A.; Sleiman, E.; Surabhi, S.; Wall, D.P. Mitigating Negative Transfer in Multi-Task Learning with Exponential Moving Average Loss Weighting Strategies. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI-23), Washington, DC, USA, 7–14 February 2023. DOI: https://doi.org/10.1609/aaai.v37i13.26983

19. Wu, Y.; Wang, X.; Wang, M.; Liu, X.; Zhu, S. Time-series forecasting of PM2.5 and PM10 concentrations based on the integration of surveillance images. Sensors 2025, 25, 95. DOI: https://doi.org/10.3390/s25010095

20. Booth, A.; James, P.; McGough, S.; Solaiman, E. Cross-regional deep learning for air quality forecasting: A comparative study of CO, NO2, O3, PM2.5, and PM10. Forecasting 2025, 7, 66. DOI: https://doi.org/10.3390/forecast7040066

21. Izonin, I.; Tkachenko, R.; Lotfi, N.; Al-kahtani, M.S. Air quality index AQI classification based on hybrid particle swarm and grey wolf optimization with ensemble machine learning model. Scientific Reports 2026, 16, 3427. DOI: https://doi.org/10.1038/s41598-025-34278-8

22. Ahmed, M.A.; Al-kahtani, M.S.; Khan, M.Z. Multimodal particulate matter prediction: Enabling scalable and high-precision air quality monitoring using mobile devices and deep learning models. Sensors 2025, 25, 4053. DOI: https://doi.org/10.3390/s25134053

23. Gorishniy, Y.; Rubachev, I.; Khrulkov, V.; Babenko, A. Revisiting Deep Learning Models for Tabular Data. In Advances in Neural Information Processing Systems (NeurIPS), Virtual Conference, 6–14 December 2021.

24. Hollmann, N.; Müller, S.; Eggensperger, K.; Hutter, F. TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second. In Proceedings of the International Conference on Learning Representations (ICLR), Kigali, Rwanda, 1–5 May 2023.

25. Zhang, Q.; Fu, F.; Tian, R. A deep learning and image-based model for air quality estimation. Science of the Total Environment 2020, 724, 138178. DOI: https://doi.org/10.1016/j.scitotenv.2020.138178