Nikos holds an MEng in Electrical and Computer Engineering from the National Technical University of Athens with specialization in computer science.

His previous research experience involves applications of deep learning techniques in computer vision problems and specifically designing deep learning methods of semi-supervised and unsupervised learning.

RESEARCH INTERESTS

Nikos’ research interests include computer vision, deep learning, causality, generative AI and medical applications. Particularly, his research will focus on developing deep learning methods which will have the ability of causal inference.

PhD research on "Counterfactual Video Generation and Evaluation"
Abstract: We propose CSVC, a framework for causally faithful counterfactual video generation using latent diffusion models. By embedding causal graphs into prompts and guiding generation with a vision-language model loss, CSVC can produce realistic, causally consistent counterfactuals. It improves causal effectiveness without fine-tuning, enabling broad applications in media and healthcare.



⚬ Recent publication: UCRL@ICLR 2026 paper "Identifiable Estimation of Causal Concept Effects under Visual Latent Confounding"