On-device User Adaptation of CNNs (Convolutional Neural Networks) 10-11 Abstract for On device User Adaptation of CNNs

Improving Utilization of NUMA (Non-Uniform Memory Access) Manycore Nodes 11-12 Abstract for Improving Utilization of NUMA manycore nodes

Bio:

Bernhard Egger is an associate professor in the Department of Computer Science and Engineering at Seoul National University, Korea, where he has been a faculty member since 2011. After receiving an engineering diploma in computer science from the Swiss Federal Institute of Technology (ETH Zurich) and a PhD in computer science and engineering from Seoul National University, he joined Samsung Electronics’ central research institution, the Samsung Advanced Institute of Technology (SAIT), where he worked on the compiler of a in-house coarse-grained reconfigurable array processor. His research interests include languages, compilers, and runtimes for distributed and parallel architectures and reconfigurable computing.