University of Moratuwa crestCSE logoComputer Science & EngineeringUniversity of Moratuwa

Trustworthy AI

Trustworthy AI

AI systems are increasingly deployed across diverse domains, including safety-critical applications such as healthcare. However, achieving high performance alone does not ensure that AI systems are trustworthy. They should also be robust, fair, interpretable, and verifiable. The Trustworthy AI Research Lab at the Department of Computer Science and Engineering, University of Moratuwa develops novel techniques for trustworthy AI and their applications in healthcare, sustainable living, and software systems through multidisciplinary research.

Selected Publications

Book Chapters

  • PH Panawenna, K G Heththihewa, S. Wickramanayake, D. Meedeniya, “Explainable AI for Building Trustworthy Transformer-Based Abstractive Summarization Models” in Explainable Artificial Intelligence for Trustworthy Decisions in Smart Applications, Springer Nature, 2025.

  •  S. Wickramanayake, S. Rasnayaka, M. Gamage, D. Meedeniya, I. Perera, “Explainable Artificial Intelligence for Enhanced Living Environments: A Study on User Perspective” in Internet of Things: Architectures for Enhanced Living Environments, Advances in Computers, 2024.

 Journal Papers

  • DY De Silva, S Wickramanayake, D Meedeniya, S Rasnayaka. SEZ-HARN: Self-explainable zero-shot human activity recognition network. Computers and Electrical Engineering. 2026

  • N Munasinghe, S Wickramanayake, and D Meedeniya. "TranSPAN: Analyzing Signal Property Learning in Transformer-Based Time Series Classification." IEEE Access (2026).

Conference Papers

  • L Akmeemana, C Attanayake, H Faiz, S. Wickramanayake, "GAL-MAD: Towards Explainable Anomaly Detection in Microservice Applications Using Graph Attention Networks", Presented at Australasian Symposium on Parallel and Distributed Computing, 2026

  • PH Panawenna, S Wickramanayake. “Understanding Omitted Facts in Transformer-based Abstractive Summarization”. In Proceeding of the Moratuwa Engineering Research Conference (MERCon) 2024.

  • S. Wickramanayake, W. Hsu and M.L. Lee, “Explanation-based data augmentation for image classification”, In Proceedings of the Advances in Neural Information Processing Systems, 2021.

  • S. Wickramanayake, W. Hsu and M.L. Lee, “Comprehensible convolutional neural networks via guided concept learning”. In 2021 International Joint Conference on Neural Networks (IJCNN), 2021.

  • S. Wickramanayake, W. Hsu. and M.L. Lee. “Flex: Faithful linguistic explanations for neural net-based model decisions”. In Proceedings of the AAAI Conference on Artificial Intelligence, 2019.

Workshop Papers

  • N N Munasinghe, and S Wickramanayake, “SPICI: Signal Property-Included Classification and Interpretability for Time-Series Transformers”. In Women in Machine Learning Workshop@ NeurIPS 2026.

  • Sandani Sesanika, and S Wickramanayake, “MetaGauge: Quantifying Metacognitive Skills in LLM Reasoning Traces”. In Women in Machine Learning Workshop@ ICML 2026.

  • Shamila Dasanayaka, and S Wickramanayake, “RobustGX: Information-Theoretic Robust Explanations for Graph Neural Networks”. In Women in Machine Learning Workshop@ ICML 2026.

  • N N Munasinghe, and S Wickramanayake, “Unveiling Signal Property Usage in Transformers for Time Series Classification”. In Women in Machine Learning Workshop@ NeurIPS 2025 

  • N Lamahewage, N Cooray, R Shariffdeen, S Wickramanayake, ND Silva. "SCHOLIA-An XAI Framework for APR" In 6th International Workshop on Automated Program Repair, 2025.

  •  S. Wickramanayake, W. Hsu, M. L. Lee. "Towards Fully Interpretable Deep Neural Networks: Are We There Yet?" In Explainable Artificial Intelligence Workshop - International Conference on Machine Learning, 2021.

Grants

  • UOM SRC Grant - SRC/ST/2025/57 for the project titled Demystifying Transformers for Time Series Forecasting. - Rs. 600,000/= Aug 2025 – Jan 2026

  • UOM SRC Grant - SRC/ST/2024/41 for the project titled Explainable artificial intelligence for building trustworthy abstractive summarization models. - Rs. 600,000/= Oct 2024 – Feb 2025

  • UOM SRC Grant - SRC/LT/2024/10 for the project titled Rationalized Artificial Intelligent Decisions through Human-Understandable Explanations for Trustworthy Internet of Things- Rs. 1,500,000/= Nov 2024 – Nov 2025

Director

Staff members

Other members