About Me

I’m an AI researcher specializing in evolving data stream learning, focusing on Neural Networks and Gradient Boosting. My work tackles key challenges in Online Learning, such as concept drift detection and mitigating catastrophic forgetting. I contribute to open-source frameworks like MOA and CapyMOA, have published in top venues, delivered tutorials, and bring industry experience in ML projects and network capture tool development. Explore my GitHub for open-source code.

Download CV
Interests
  • Stream Learning
  • Online Continual Learning
  • Online Streaming Continual Learning
Education

  • PhD Artificial Intelligence

    University of Waikato

  • MSc in Information Sciences

    Auckland University of Technology

  • BSc in Management Information Systems

    University College Dublin

📚 My Research
I am an AI researcher specializing in evolving data stream learning, focusing on advanced techniques like Neural Networks and Gradient Boosting. My work addresses critical challenges in Online Learning, including detecting and adapting to concept drifts in data streams and mitigating catastrophic forgetting in Neural Networks for online learning. I actively contribute to the open-source MOA and CapyMOA stream learning frameworks. I have also published multiple publications in reputed venues and have delivered tutorials on online learning. I have extensive experience in the industry, having participated in multiple machine learning projects and developed network capture tools. Explore my GitHub for accessible source codes.
Projects
FREEWAY
FREEWAY

develops an Asynchronous Federated Learning framework to improve vehicle operation analytics for diverse fleets of commercial EVs, tackling scalability, connectivity, and efficiency. By combining edge computing and MLOps, it focuses on real-time energy forecasting, activity recognition, and anomaly detection while ensuring data privacy. The consortium includes Volvo Trucks, Swedish research institutes, and SMEs.

Jul 28, 2025

AIM-TRUE
AIM-TRUE

uses AI to optimize Volvo’s aftermarket services, boosting efficiency and part availability via predictive logistics.

Jul 27, 2025

KEEPER
KEEPER

transforms industrial data into actionable insights using AI, to optimize assets like trucks, pumps, and network equipment, collaborating with Swedish industry partners and research institutes.

Jul 26, 2025

CapyMOA
CapyMOA

Machine learning library tailored for data streams. Featuring a Python API tightly integrated with MOA (Stream Learners), PyTorch (Neural Networks), and scikit-learn (Machine Learning). CapyMOA provides a fast python interface to leverage the state-of-the-art algorithms in the field of data streams.

Oct 26, 2023

MOA
MOA

MOA is the most popular open source framework for data stream mining.

Oct 26, 2023

Featured Publications
Recent Publications
Nuwan Gunasekara, Bernhard Pfahringer, Heitor Gomes, Albert Bifet (2025). Gradient boosted bagging for evolving data stream regression. Data Mining and Knowledge Discovery.
Cite
Heitor Murilo Gomes, Anton Lee, Nuwan Gunasekara, Yibin Sun, Guilherme Weigert Cassales, Justin Liu, Marco Heyden, Vitor Cerqueira, Maroua Bahri, Yun Sing Koh, Others (2025). CapyMOA: Efficient Machine Learning for Data Streams in Python. arXiv preprint arXiv:2502.07432.
Cite
Nuwan Gunasekara, Sławomir Nowaczyk, Sepideh Pashami (2025). Pragmatic Paradigm for Multi-stream Regression. International Symposium on Intelligent Data Analysis.
Cite
Nuwan Gunasekara, Bernhard Pfahringer, Heitor Gomes, Albert Bifet (2024). Gradient boosted trees for evolving data streams. Machine Learning.
Cite
Nuwan Gunasekara, Bernhard Pfahringer, Heitor Murilo Gomes, Albert Bifet, Yun Sing (2024). Recurrent concept drifts on data streams. Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24.
Cite
See all publications
Tutorials