Visiting address
DS - BI
University of Groningen
Nijenborgh 9
9747 AG Groningen
The Netherlands
Quick links
Follow us
TODO
Bachelor Projects
Automated Dataset Generator for Wastewater System Simulations
Supervisors:
Dilek Düştegör,
Revin Alief
Date: 2025-01-21
Type: bachelor
Description:
Automating dataset generation is essential for streamlining the preprocessing pipeline, enabling faster iterations and ensuring scalability as wastewater systems become increasingly complex. This project involves designing and implementing a program to automate dataset generation for wastewater simulations. The program will take inputs such as simulation parameters, geographic data, and network structures (e.g., pipe diameters and node connections) and output datasets compatible with Graph Neural Network (GNN) models. The student will focus on creating a flexible, user-friendly interface for defining input parameters, ensuring compatibility with graph-based models and simulation outputs, and testing the tool across multiple simulation scenarios. The expected outcome is a reusable dataset generation tool that significantly enhances the efficiency of wastewater simulation workflows.
References:
Infoworks ICM Exchange Infoworks Ruby Scripts
References:
Infoworks ICM Exchange Infoworks Ruby Scripts
Optimizing Graph Neural Networks for Water Level Estimation
Supervisors:
Dilek Düştegör,
Revin Alief
Date: 2025-01-21
Type: bachelor
Description:
Optimizing Graph Neural Networks (GNNs) is critical for enhancing the accuracy and efficiency of water level predictions, which directly influences the reliability of wastewater management systems. This project focuses on refining specific aspects of GNNs for predicting water levels at nodes in wastewater networks. The student will explore hyperparameter tuning, feature engineering, and advanced GNN variants such as Graph Attention Networks (GAT). Key tasks include studying the impact of different GNN architectures and parameters on performance, experimenting with incorporating edge features (e.g., pipe diameters), and evaluating models on datasets generated by ICM. The expected outcome is a comprehensive analysis of optimization techniques for GNNs in the wastewater domain.
References:
Zhang, Z., Tian, W., Lu, C., Liao, Z., & Yuan, Z. (2024). Graph neural network-based surrogate modelling for real-time hydraulic prediction of urban drainage networks. Water Research, 263, 122142. https://doi.org/10.1016/j.watres.2024.122142 Li, M., Shi, X., Lu, Z., & Kapelan, Z. (2024). Predicting the urban stormwater drainage system state using the Graph-WaveNet. Sustainable Cities and Society, 115, 105877. https://doi.org/10.1016/j.scs.2024.105877
References:
Zhang, Z., Tian, W., Lu, C., Liao, Z., & Yuan, Z. (2024). Graph neural network-based surrogate modelling for real-time hydraulic prediction of urban drainage networks. Water Research, 263, 122142. https://doi.org/10.1016/j.watres.2024.122142 Li, M., Shi, X., Lu, Z., & Kapelan, Z. (2024). Predicting the urban stormwater drainage system state using the Graph-WaveNet. Sustainable Cities and Society, 115, 105877. https://doi.org/10.1016/j.scs.2024.105877
Benchmarking AI Workloads on GPU Cluster
Supervisors:
Kawsar Haghshenas,
Mahmoud Alasmar
Date: 2025-01-21
Type: bachelor
Description:
Understanding the characteristics of AI workloads is essential for effective resource allocation and fault tolerance mechanisms. This project focuses on benchmarking various deep neural network (DNN) models on GPUs using different profiling and monitoring tools. You will observe and analyze their runtime behavior, identify the factors affecting model performance, and propose metrics that effectively quantify their runtime characteristics. The outcome of this project is to deliver a comprehensive study on profiling DNN models with minimal overhead and maximum accuracy.
References:
Gao, Wanling, et al. "Data motifs: A lens towards fully understanding big data and ai workloads." Proceedings of the 27th International Conference on Parallel Architectures and Compilation Techniques. 2018., https://arxiv.org/abs/1808.08512 Xiao, Wencong, et al. "Gandiva: Introspective cluster scheduling for deep learning." 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18). 2018., https://www.usenix.org/conference/osdi18/presentation/xiao Yang, Charlene, et al. "Hierarchical roofline performance analysis for deep learning applications." Intelligent Computing: Proceedings of the 2021 Computing Conference, Volume 2. Springer International Publishing, 2021., https://arxiv.org/abs/2009.05257
References:
Gao, Wanling, et al. "Data motifs: A lens towards fully understanding big data and ai workloads." Proceedings of the 27th International Conference on Parallel Architectures and Compilation Techniques. 2018., https://arxiv.org/abs/1808.08512 Xiao, Wencong, et al. "Gandiva: Introspective cluster scheduling for deep learning." 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18). 2018., https://www.usenix.org/conference/osdi18/presentation/xiao Yang, Charlene, et al. "Hierarchical roofline performance analysis for deep learning applications." Intelligent Computing: Proceedings of the 2021 Computing Conference, Volume 2. Springer International Publishing, 2021., https://arxiv.org/abs/2009.05257
DiTEC project- Unsupervised Learning for Customer Profiles in Water Distribution Networks
Supervisors:
Huy Truong,
Dilek Düştegör
Date: 2025-01-21
Type: bachelor
Description:
Researchers studying drinking water distribution networks often rely on large-scale synthesized datasets. However, the current simulation generating these datasets faces limitations in retrieving metadata that enhances dataset accessibility. Moreover, this missing metadata, including customer profiles at each node in the network, plays a crucial role in classifying customer types and estimating their demand, particularly during peak seasons. To address this gap, the student could apply unsupervised clustering algorithms such as K-NN, K-means, or DBSCAN to identify and retrieve these customer profiles. This project requires a candidate with a solid background in Machine Learning and interest in building robust data pipelines. They will be eventually employed to extract the missing metadata for a large-scale dataset, enabling water experts to analyze water network and benchmark customer behavior efficiently.
References:
Tello, A., Truong, H., Lazovik, A., & Degeler, V. (2024). Large-scale multipurpose benchmark datasets for assessing data-driven deep learning approaches for water distribution networks. Engineering Proceedings, 69(1), 50. https://doi.org/10.3390/engproc2024069050.
References:
Tello, A., Truong, H., Lazovik, A., & Degeler, V. (2024). Large-scale multipurpose benchmark datasets for assessing data-driven deep learning approaches for water distribution networks. Engineering Proceedings, 69(1), 50. https://doi.org/10.3390/engproc2024069050.
Node masking in Graph Neural Networks
Supervisors:
Huy Truong,
Dilek Düstegör
Date: 2025-01-21
Type: bachelor
Description:
Working with data in the real world often leads to missing information problems, which can negatively affect the performance of deep learning models. However, in proper ways, it can boost the expressiveness of Graph Neural Network (GNN) models in node representation learning through a technique known as Node Masking. In particular, it hides arbitrary nodal features in a graph and instructs the GNN to recover the missing parts. The student can explore diverse masking strategies, such as zero masking, random node replacement, mean-neighbor substitution, shared learnable embedding, and nodal permutation. These options above should be compared and evaluated in a graph reconstruction task that applies to a water distribution network. This study will focus on finding a generative technique that effectively enhances the performance of GNN models in semi-supervised transductive learning. Students interested in joining this project should possess a machine-learning background and a deep-learning framework.
References:
Hou, Zhenyu, Xiao Liu, Yuxiao Dong, Chunjie Wang, and Jie Tang. "GraphMAE: Self-Supervised Masked Graph Autoencoders." arXiv preprint arXiv:2205.10803(2022). Abboud, Ralph, Ismail Ilkan Ceylan, Martin Grohe, and Thomas Lukasiewicz. "The surprising power of graph neural networks with random node initialization." arXiv preprint arXiv:2010.01179 (2020). Hajgató, Gergely, Bálint Gyires-Tóth, and György Paál. "Reconstructing nodal pressures in water distribution systems with graph neural networks." arXiv preprint arXiv:2104.13619 (2021). He, Kaiming, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. "Masked autoencoders are scalable vision learners." In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 16000-16009. 2022.
References:
Hou, Zhenyu, Xiao Liu, Yuxiao Dong, Chunjie Wang, and Jie Tang. "GraphMAE: Self-Supervised Masked Graph Autoencoders." arXiv preprint arXiv:2205.10803(2022). Abboud, Ralph, Ismail Ilkan Ceylan, Martin Grohe, and Thomas Lukasiewicz. "The surprising power of graph neural networks with random node initialization." arXiv preprint arXiv:2010.01179 (2020). Hajgató, Gergely, Bálint Gyires-Tóth, and György Paál. "Reconstructing nodal pressures in water distribution systems with graph neural networks." arXiv preprint arXiv:2104.13619 (2021). He, Kaiming, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. "Masked autoencoders are scalable vision learners." In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 16000-16009. 2022.