Parham Zilouchian Moghaddam

I am a machine learning researcher specializing in the development of innovative and practical artificial intelligence solutions at the intersection of deep learning, generative models, and computer vision. As the top-ranked graduate of the Master of Science program in Computer Engineering at the University of Tehran, my work is dedicated to addressing complex, real-world challenges, including the enhancement of video compression, the improvement of speech processing, and the advancement of trustworthy AI.

My Master’s research, conducted under the esteemed mentorship of Dr. Mehdi Modarressi and Dr. Mohammad Amin Sadeghi, culminated in a thesis focused on designing generative models for video compression. A significant outcome of this research was the development of “NU-Class Net”, a novel deep neural network aimed at substantially enhancing the quality of low-bitrate video. This work not only demonstrated a considerable reduction in video size but also led to a journal paper submitted to Elsevier’s Engineering Applications of Artificial Intelligence. Additionally, my research experience encompasses pioneering efforts in generative networks for voice adaptation and optimizing the efficiency of Generative Adversarial Networks (GANs) and Long Short-Term Memory (LSTM) networks.

Currently, I am applying my expertise as a researcher at the Qatar Computing Research Institute (QCRI), where I contribute to Fanar, a large-scale generative AI initiative. In this role, I am specifically responsible for benchmarking the image generation capabilities of this multi-modal project. My commitment to bridging the gap between research and practical applications is further exemplified by my experience as a computer vision developer and as a co-founder of a startup. I am dedicated to advancing the field and actively share my knowledge through technical writing and content creation. I welcome opportunities to connect with fellow researchers and industry professionals to explore collaborative endeavors and address the next generation of challenges in artificial intelligence.