HKBU and Tencent break world record for fast and accurate AI training
06 Nov 2018
HKBU researchers have partnered with a team from Tencent Machine Learning to create a new technique for training Artificial Intelligence (AI) machines faster than ever before, while maintaining accuracy. The HKBU team is comprised of Professor Chu Xiaowen and PhD student Shi Shaohuai from the Department of Computer Science.
During the experiment the team succeeded in training two popular deep neural networks called AlexNet and ResNet-50 in just 4 minutes and 6.6 minutes respectively. Previously the fastest time was 11 minutes for AlexNet and 15 minutes for ResNet-50.
AlexNet and ResNet-50 are deep neural networks built on ImageNet, a large-scale dataset for visual recognition. Once trained, the system was able to recognise and label an object in a given photo. The result is significantly faster than previous records and outperforms all other existing systems.
For more details, please refer to: https://cpro.hkbu.edu.hk/en/press_release/detail/HKBU-and-Tencent-break-world-record-for-fast-and-accurate-AI-training/.