Modular AI Verification and Visualisation (MAIVV)
(2021-2026)
(2021-2026)
Modular AI Verification and Visualisation (MAIVV) is led by Professor Rosemary Monahan and Professor Barak A. Pearlmutter in the Dept. of Computer Science and Hamilton Institute at Maynooth University. The project is funded from December 2021 until November 2026 by Taighde Éireann - Research Ireland under grant 20/FFP-P/8853, and the project team includes postdoctoral researchers Medet Inkarbekov and Syed Ali Asadullah Bukhari and PhD student Thomas Flinkow.
Our overall goal is to provide scalable techniques for software development that guarantee software dependability.
Highlights
Some information on our research follows at the links below:
- NeuralVizVR: An Open source platform to improve neural network interpretability and analysis by leveraging immersive 3D visualization, interactive exploration, and advanced AI-assisted techniques, making complex models more accessible and understandable to a wide range of users.
- Comparing Differentiable Logics for Learning with Logical Constraints: A promising approach for creating machine learning models that inherently satisfy constraints after training is to encode background knowledge as explicit logical constraints that guide the learning process via so-called differentiable logics. Here, we experimentally compare and evaluate various logics from the literature, presenting our findings and highlighting open problems for future work.
- Creating a Formally Verified Neural Network for Autonomous Navigation: We present our experience of a case study exploring the design and training of a neural network on a custom dataset for vision-based autonomous navigation.
Publications
- Verification-Aware Convolution Neural Networks for Speech Recognition: A case study Formal Methods in Software Engineering (Formalise'26) 2026 DOI: 10.1145/3793656.3793683
- Comparing differentiable logics for learning with logical constraints Science of Computer Programming 2025 DOI: 10.1016/j.scico.2025.103280
- A General Framework for Property-Driven Machine Learning arXiv preprint 2025 arXiv:2505.00466v2
- Creating a Formally Verified Neural Network for Autonomous Navigation: An Experience Report FMAS'24 2024 DOI: 10.4204/EPTCS.411.12
- Towards Correct-by-Construction Machine-Learnt Models Doctoral Forum at iFM'24 2023 Link
- Immersive Neural Network Exploration: A VR Approach to Human-Centered AI Understanding Human Centered AI Education and Practice Conference 2023 (HCAIep'23) 2023 DOI: 10.1145/3633083.3633221
- Comparing Differentiable Logics for Learning Systems: A Research Preview FMAS'23 2023 DOI: 10.4204/EPTCS.395.3
- Visualization of AI Systems in Virtual Reality: A Comprehensive Review (IJACSA) International Journal of Advanced Computer Science and Applications 2023 Link
- Rich and Expressive Specification of Continuous-Learning Cyber-Physical Systems Doctoral Forum at DSN'23 2023 DOI: 10.1109/DSN-S58398.2023.00054
Presentations
Talks
- Machine Learning with Verifiable Guarantees VerifAI'26 2026 Slides
- Differentiable Logic for Correct-by-Construction Neural Networks Dagstuhl Seminar 26031: Software Contracts Meet System Contracts 2026 Slides
- A Framework for Property-driven Machine Learning in PyTorch (Invited) Heriot-Watt University Edinburgh 2025 Slides
- Property-driven Machine Learning Maynooth University CS Postgraduate Workshop 2025 2025 Slides
- Towards Correct-by-Construction Machine-Learnt Models Doctoral Symposium at iFM'24 2024 Slides
- Creating a Formally Verified Neural Network for Autonomous Navigation: An Experience Report FMAS'24 2024 Slides
- Property-driven Machine Learning with Differentiable Logics (Invited) Heriot-Watt University Edinburgh 2024 Slides
- Differentiable Logics for Machine Learning with Logical Constraints in Practice (Invited) IT University Copenhagen 2024 Slides
- Differentiable Logics for Learning with Logical Constraints VTSA 2024 Slides
- Comparing Differentiable Logics for Learning Systems (A Research Preview) FMAS 2023 2023 Slides
Posters
- Towards Verification-Aware Neural Networks for Speech Recognition Maynooth University CS Research Week 2025 Poster
- Property-driven Machine Learning with Differentiable Logics Maynooth University CS Research Week 2025 Poster
- Comparing Differentiable Logics for Learning with Logical Constraints SAIV'25' 2025 Poster
- Towards Correct-by-Construction Machine Learning via Differentiable Logics ADAPT Conference 2025 Poster
- Creating a Formally Verified Neural Network for Autonomous Navigation Maynooth University CS Research Week 2024 Poster
- Differentiable Logics for Machine Learning: What difference do they make? iFM Conference 2024 Poster
- Machine Learning with Differentiable Logics Maynooth University CS Research Week 2023 Poster