I am a PhD student jointly at IRLab and PCS research groups at the University of Amsterdam. I am working on "Efficiency and scalability of search recommendation ", and  I am supervised by Andrew Yates and Ana-Lucia Varbanescu. 

I completed my Master's degree in digital electronic systems at Amirkabir Univeristy of Tehchnology, Tehran, Iran in 2020.  My Master's thesis was about Persian language modeling using transformers. Before my master's, I finished my Bacheolors in electrical engineering at Amirkabir Univeristy of Technology in 2017.

Research

I am passionate about building tools and systems and enjoy collaborating with a large team on large-scale software projects. In particular, I am interested in the efficiency and scalability of large-scale problems in AI, particularly search and recommendation. Either through data and algorithmic efficiency by dataset sampling, dataset distillation, and model distillation or by performance analysis and applying parallel computing techniques.

Distillation vs. Sampling for Efficient Training of Learning to Rank Models

Pooya Khandel, Andrew Yates, Ana-Lucia Varbanescu, Maarten de Rijke, Andy Pimentel

 ICTIR'24 / Code / Paper

MassiveClicks: A Massively-parallel Framework for Efficient  Click Models Training

Skip Thijssen, Pooya Khandel, Andrew Yates, Ana-Lucia Varbanescu

 EuroPar'23: Hetropar Workshop / Paper

ParClick: A Scalable Algorithm for EM-based Click Models

Pooya Khandel, Ilya Markov, Andrew Yates, Ana-Lucia Varbanescu

 WebConf'22 / Code / Paper

SensorDrop: A Reinforcement Learning Framework for Communication Overhead Reduction on the

Pooya Khandel, Amir Hossein Rassafi , Vahid Pourahmadi , Saeed Sharifian , and Rong Zheng

 arxiv 2019 / Paper

Thanks to Jon for template design!