Primarily, my work falls under the umbrella of Artificial Intelligence. I design and analyze algorithms that learn, with a focus on Algorithmic Game-Theoretic Learning - an interdisciplinary field that synthesizes algorithmic and game-theoretic perspectives to develop learning algorithms for competitive and random environments.

I am particularly interested in online learning algorithms that adapt and learn sequentially from streaming data, with a focus on zero-sum repeated games and adversarial learning scenarios. My research examines the dynamic interplay between multiple decision-makers in strategic interactions, developing solutions that exhibit provably desirable predictive, space, and time complexities while maintaining strategic robustness.

This interdisciplinary approach enables the creation of adaptive systems capable of principled decision-making. My research contributes to the fundamental understanding of how intelligent systems can operate effectively in environments with multiple competing agents while ensuring both computational efficiency and strategic soundness.

PROTEUS (Completed)

European research project designed to address fundamental scientific challenges related to the scalability and responsiveness of analytics capabilities. Focused on developing advanced data analytics solutions for complex, large-scale systems. This project contributed to the advancement of scalable machine learning algorithms for real-time data processing.
Details

ETI - Energy Technologies Institute (Completed)

High Frequency Appliance Disaggregation Analysis project analyzing real-world data from ETI's Home Energy Management System in five homes to gather detailed energy data from water, gas and electricity use. This research focused on energy consumption pattern analysis and predictive modeling for smart home applications.
Details

ExtremeXP (Active)

Aims to provide accurate, precise, fit-for-purpose, and trustworthy data-driven insights via evaluating different complex analytics variants, considering end users preferences and feedback in an automated way. Includes binary classification research for malicious network traffic detection with online learning algorithms. This project demonstrates the practical application of algorithmic game-theoretic learning in cybersecurity contexts.
Details

• Jamil, W. and Bouchachia, A. (2022). Iterative ridge regression using the aggregating algorithm. Pattern Recognition Letters 158, 34-41.
• Jamil, W. and Bouchachia, A. (2020). Competitive normalized least-squares regression. IEEE Transactions on Neural Networks and Learning Systems 32 (7), 3262-3267.
• Jamil, W. and Bouchachia, A. (2020). Competitive regularised regression. Neurocomputing 390, 374-383.
• Jamil, W. and Bouchachia, A. (2020). Online Bayesian shrinkage regression. Neural Computing and Applications 32 (23), 17759-17767.
• Jamil, W. and Bouchachia, A. (2018). Model selection in online learning for times series forecasting. Proceedings of the 18th Annual UK Workshop on Computational Intelligence, 83-95.
• Jamil, W., Duong, N.C., Wang, W., Mansouri, C., Mohamad, S. and Bouchachia, A. (2018). Scalable online learning for flink: SOLMA library. Proceedings of the 12th European Conference on Software Architecture.
• Jamil, W., Kalnishkan, Y. and Bouchachia, A. (2016). Aggregation algorithm vs. average for time series prediction. Proceedings of the ECML PKDD 2016 Workshop on Large-scale Learning from Data Streams in Evolving Environments.