Muhammad Ajmal Naz

Ph.D. candidate in Ingegneria Elettrica, Elettronica E Delle Comunicazioni , 41^st cycle (2025-2028)
Department of Electronics and Telecommunications (DET)

muhammad.naz@polito.it

Profile

PhD

Research topic

Multimodal Deep Learning for Conversational Response Generation

Tutors

Ladislau Matekovits
Matej Rojc

Keywords

Big Data, Machine Learning, Neural Networks and Data Science

Biography

PhD research on generating contextually aware and natural conversational responses through deep learning-based fusion of verbal and non-verbal signals. Develops low and high level fusion algorithms trained on multimodal conversational corpora to classify communicative intent and predict dialogue acts that reflect conversational context, emotional and gestural information. Applied to both closed- and open-domain dialogue systems; evaluation combines quantitative metrics (intent classification accuracy) with qualitative user studies assessing response naturalness and appropriateness.