Semi-Supervised Distillation: Personalizing Deep Neural Networks in Activity Recognition using Inertial Sensors
Yusuke Iwasawa, Ikuko Eguchi Yairi, Yutaka Matsuo · Transactions of the Japanese Society for Artificial Intelligence · 2017
Personalization of activity recognition has become a topic of interest to improve recognition performance for diverse users. Recent researches show that deep neural networks improve generalization performance in activity recognition using inertial sensors, such as accelerometers and gyroscopes; however, personalizing deep neural networks is challenging because it has a thousands or millions of parameters but generally personalization should be done with small amount of labeled data.