Greetings from Vienna

Jonathan Svirsky, a PhD student of Dr. Ofir Lindenbaum’s, visited Vienna in July to attend the International Conference on Machine Learning (ICML). “It’s a highly-rated annual conference that takes place somewhere different each year. Every year, more than 1,000 research papers apply, but only 25% are accepted.” He tells. “The conference spans six days, and we presented a poster that attracted much attention.”

Svirsky, 36, married +5, is a third-year doctoral student in the electrical engineering program. He earned his bachelor’s and master’s degrees at the Technion, before arriving at our Faculty of Engineering to work with Dr. Lindenbaum. “Ofir and I met several years ago, and we get together every week. I recommend him to every student—a researcher with vast knowledge in a myriad of fields, a professional and dedicated supervisor with incredible people skills.”

The two presented their research on sparsity, the representation of information in neural networks.” Sparsity allows us to weed out non-vital information in exchange for a more efficient and precise representation of data in neural networks. Our research explores audio, images, and tabular data. At this conference, we presented a project concerning deep clustering with interpretability, made possible via the sparsification of the samples in their original space,” he explains. “Reactions to the poster were very positive, we exchanged some ideas with researchers in the field, and most importantly—we received great feedback and ideas for future research based on the work we presented.”