Artificial Intelligence News

We're happy to announce that Shuichi Hirahara has joined the department as a Research Fellow.

Shuichi completed his PhD at the University of Tokyo in 2019. He is currently an Associate Professor at the National Institute of Informatics, Tokyo.

Shuichi's primary research area is computational complexity theory. During his stay at Warwick, he will be involved in the activities of a joint project with the University of Oxford on the limits and possibilities of efficient algorithms.

We are happy to announce that Torsten Mütze (left in the picture), assistant professor in the Theory and Foundations Research Division, has won the Best Paper Award at the 47th International Symposium on Mathematical Foundations of Computer Science (MFCS 2022) for the paper "The Hamilton compression of highly symmetric graphs", authored jointly with his student Arturo Merino (TU Berlin; middle) and Petr Gregor (Charles University Prague; right). The paper proposes a new graph parameter that measures the amount of symmetry present in its Hamilton cycles, and it investigates this parameter for a wide range of interesting highly-symmetric graphs. It combines methods from combinatorics, number theory and algebra, and connects the new parameter to several related problems that researchers have studied intensively. The MFCS best paper award is sponsored by the European Association for Theoretical Computer Science.

Our teams from Warwick DCS have won both the 1st and 2nd place at the Zero Cost NAS Competition held in conjunction with the AutoML'22 conference.

Over the recent year, Neural Architecture Search (NAS) has attracted a lot of attention. While being able to automate the discovery of better performing neural architectures than hand-crafted ones, it comes at a great price, requiring thousands of GPU hours to perform the search. The Zero Cost NAS competition challenges the participants to design efficient proxies for NAS, using negligible computational resources to evaluate neural architectures.

In collaboration with the AutoCAML team at Samsung AI Cambridge (led by Dr. Hongkai Wen), our research students, Lichuan Xiang and Youyang Sha, proposed new zero-cost NAS metrics that exploit the compressibility of neural networks. Our metrics are extremely efficient to run (reducing search cost from weeks/days to minutes), and achieves impressive results across multiple search spaces and datasets. In the competition, our teams won both the 1st and 2nd places (using different scoring functions), and the performance gap with the 3rd winning team is almost 2x. Checkout our poster here.