AI Foundational Optimizations

This project focuses on advancing the efficiency, scalability, and reliability of core AI models and algorithms. By optimizing training methods, model architectures, and computational processes, it drives the development of robust and high-performance AI systems capable of adapting to dynamic and data-intensive tasks.

People

Dr. Sanjay Chawla

Chief Scientist

View Profile

Dr. Mohamed M. Hefeeda

Research Director, QCAI

View Profile

Dr. Anas A. Al-Nuaimi

Senior Scientist

View Profile

Dr. Michaël J. Aupetit

Senior Scientist

View Profile

Dr. Abdelkader Baggag

Senior Scientist

View Profile

Dr. Mohammad Amin Sadeghi

Senior Scientist

View Profile

Dr. Safa Messaoud

Scientist

View Profile

Keivin Isufaj

Software Engineer

View Profile

Johanne G. Medina

Research Associate

View Profile

Igli Mlloja

Research Assistant

View Profile

Publications

Rapid disaster damage assessment using deep adversarial sliced Wasserstein domain adaptation

Fatma AlNaimi , Abdulaziz Al-Homaid , Ferda Ofli , Abdelkader Baggag
Neural Computing and Applications (2026)

Particles Don’t Care About Z: Towards Scaling Entropy Estimation of Unnormalized Densities

Safa Messaoud, Skander Charni, Elaa Bouazza, Ali Pourghasemi, Halima Bensmail
ICML (2026)

Uncertainty-Aware LLMs Fail to Flag Misleading Contexts

Tianyi Zhou, Johanne Medina, Sanjay Chawla
NeurIPS 2025 – Reliable ML Workshop (2025)

Explaining the role of Intrinsic Dimensionality in Adversarial Training

Enes Altinisik, Safa Messaoud, Husrev Taha Sencar, Hassan Sajjad, Sanjay Chawla
ICML (2025)

Deep learning, transformers and graph neural networks: a linear algebra perspective

Abdelkader Baggag, Yousef Saad
Numerical Algorithms (2025)

Projects and Solutions

Feedback-guided Self-Improving Model for Fanar Arabic GenAI using MLOps and Taxonomy Evolution

Feedback-guided Self-Improving Model for Fanar Arabic GenAI using MLOps and Taxonomy Evolution

This work builds on top of a fundamental conviction that the development of production-grade ML-based systems starts after the first deployment of an initial model.…