I am a third-year PhD student in the Computer Lab, University of Cambridge, supervised by Dr Damon Wischik. My main research interest is in developing generative models for unstructured data. I study latent-space disentanglement for representing grouped data and solving missing value problems. I am funded by the Alan Turing Institute, UK’s national AI research institute.

I have previously obtained an MPhil in Advanced Computer Science at the University of Cambridge, and a BSc in Computer Science at the University of Manchester.

Publications

1. D.A. Iliescu, D.J. Wischik. Group Disentanglement Under Conditional Shift, under review.
2. A. Mustafa, A. Mikhailiuk, D.A. Iliescu, V. Babbar, R.K. Mantiuk. Training a Task-Specific Image Reconstruction, Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2022, pp. 2319-2328.
3. D.A. Iliescu, A. Mikhailiuk, D.J. Wischik, R.K. Mantiuk. Disentangling Domain and Content. arXiv:2202.07285, 2020.
4. A. Webb, C. Reynolds, D.A. Iliescu, W. Chen, H. Reeve, M. Luján, G. Brown. To Ensemble or Not Ensemble: When Does End-to-End Training Fail?, Machine Learning and Knowledge Discovery in Databases, 2021, pp.109-123.
5. D.A. Iliescu, F. Petrogalli. Arm Scalable Vector Extensions and Application to Machine Learning, Whitepaper, ARM Ltd, 2018.

Teaching

I supervise Cambridge undergraduate students in the following courses:

• Information Theory, Part II, Spring 2022
• Machine Learning and Bayesian Inference, Part II, Spring 2020, Spring 2021, Spring 2022
• Data Science, Part IB, Autumn 2021
• Economics, Law and Ethics, Part IA, Autumn 2019

Art

I also enjoy painting and developing ML tools for aiding artists. You can check out some of my work on Behance. My work has recently won an art competition at St Edmund’s College, Cambridge.