FocusLearn: Fully-Interpretable, High-Performance Modular Neural Networks for Time Series
Su, Q.; Kloukinas, C.; Garcez, A (2024) "FocusLearn: Fully-Interpretable, High-Performance Modular Neural Networks for Time Series." arxiv.org/abs/2311.16834
CV
Education
- PhD in Explainable AI, City, University of London, London, 2020-2024 (expected)
- First Supervisor: Dr. Christos Kloukinas, Reader, City, University of London
- Second Supervisor: Prof. Artur d’Garcez, Professor, City, University of London
- MSc in Data Science, City, University of London, London, 2019-2020
- BSc in Mathematics, University of Surrey, Guildford, 2015-2019
Work experience
Machine Learning Enginner, Ofcom, 2024-Present
Algorithimic assessment for Online Safety.
- Leading a team of four in a pilot research for online services compliance with the Online Safety Act.
- Developed assessment metrics for online services to demonstrate compliance to the Online Safety Act.
- Consulted on regulatory policies regarding Online Safety Act.
- Advised the Secretary of States on assessing recommender system algorithms and mitigating potential online harms.
- Communicated research findings to diverse audiences.
- Contributing to guidelines on explainability for major social media platforms to align with Online Safety Act regulations.
- Serving as Scrum Master, overseeing project and sprint planning.
AI Researcher, SMART BEAR EU Project, 2020-2024
Researcher for SMART BEAR developing a big data platform that will provide evidence-based personalised support for several pressing health-care issues faced by the aging EU societies.
- I led the XAI-driven research as part of my PhD study to personalise health interventions and promote independent living for elderly patients from six EU countries
- Design a data-driven pipeline for Hearing Loss patients that provides clinicians with decision-making support by leveraging AI and XAI techniques. Further improved collaboration skills while working in a large and complex cross-functional team.
- Spearheaded project to be tested with 5,000 elderly patients by 2023.
Graduate Teaching Assistant, City, University of London, 2020-2024
- Assisted in teaching practical labs and marking coursework/thesis for undergraduate and postgraduate Computer Science students.
- Delivered modules in Data Science, Machine Learning, Big Data, Introduction to Java, and Theory of Computation
- Improved students’ learning and understanding of key concepts and technologies in Computer Science
Publications
The SMART BEAR Project: An Overview of Its Infrastructure
Su, Q.; Peretokin, V.; Basdekis, I.; Kouris, I.; Maggesi, J.; Sicuranza, M.; Acebes, A.; Bucur, A.; Mukkala, V.; Pozdniakov, K.; Kloukinas, C.; Koutsouris, D.; Iliadou, E.; Leontsinis, I.; Gallo, L.; De Pietro, G. and Spanoudakis (2023) "The SMART BEAR Project: An Overview of Its Infrastructure." Information and Communication Technologies for Ageing Well and e-Health. ICT4AWE ICT4AWE 2021 2022. Communications in Computer and Information Science, vol 1856. Springer, Cham.
Predicting and Explaining Hearing Aid Usage Using Encoder-Decoder with Attention Mechanism and SHAP
Su, Q. and Iliadou, E.(2022). "Predicting and Explaining Hearing Aid Usage Using Encoder-Decoder with Attention Mechanism and SHAP." In 2022 16th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS) (pp. 308-315). IEEE Computer Society.
Profiling Hearing Aid Users through Big Data Explainable Artificial Intelligence Techniques
Iliadou, E., Su, Q., Kikidis, D., Bibas, T. and Kloukinas, C. (2022) "Profiling hearing aid users through big data explainable artificial intelligence techniques." Frontiers in Neurology, 13.
Overview of the SMART-BEAR Technical Infrastructure
Peretokin, V.; Basdekis, I.; Kouris, I.; Maggesi, J.; Sicuranza, M.; Su, Q.; Acebes, A.; Bucur, A.; Mukkala, V.; Pozdniakov, K.; Kloukinas, C.; Koutsouris, D.; Iliadou, E.; Leontsinis, I.; Gallo, L.; De Pietro, G. and Spanoudakis, G. (2022). "Overview of the SMART-BEAR Technical Infrastructure." In Proceedings of the 8th International Conference on Information and Communication Technologies for Ageing Well and e-Health - ICT4AWE. Pages 117-125.
Skills
- Explainable AI
- Deep Neural Networks
- Machine Learning
- NLP
- Time Series
- Big Data
- Python
- MatLab
- Java
- TensorFlow
- PyTorch
- Scikit-learn
- PySpark
Languages
- Chinese: Native (Verbal & Written)
- English: Fluent (Verbal & Written)
- Spanish: Conversational
Certifications
- Piano: ABRSM - Grade 5 & Central Conservatory of Music (China) - Grade 7
- HE Teaching: Associate Fellow