Dual-Memory Deep Learning Architectures for Lifelong Learning of Everyday Human Behaviors
Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence. Pages 1669-1674.
Abstract
Learning from human behaviors in the real world is important for building human-aware intelligent systems such as personalized digital assistants and autonomous humanoid robots.Everyday activities of human life can now be measured through wearable sensors.However, innovations are required to learn these sensory data in an online incremental manner over an extended period of time. Here we propose a dual memory architecture that processes slow-changing global patterns as well as keeps track of fast-changing local behaviors over a lifetime. The lifelong learnability is achieved by developing new techniques, such as weight transfer and an online learning algorithm with incremental features. The proposed model outperformed other comparable methods on two real-life data-sets: the image-stream dataset and the real-world lifelogs collected through the Google Glass for 46 days.
@inproceedings{ijcai2016p239,
title = {Dual-Memory Deep Learning Architectures for Lifelong Learning of Everyday Human Behaviors},
author = {Lee, Sang-Woo and Lee, Chung-Yeon and Kwak, Dong-Hyun and Kim, Jiwon and Kim, Jeonghee and Zhang, Byoung-Tak},
booktitle = {Proceedings of the Twenty-Fifth International Joint Conference on
Artificial Intelligence, {IJCAI-16}},
publisher = {AAAI Press / International Joint Conferences on Artificial Intelligence},
editor = {Subbarao Kambhampati},
pages = {1669--1674},
year = {2016},
month = {7},
url = {https://ijcai.org/proceedings/2016/239},
}
TY - CPAPER TI - Dual-Memory Deep Learning Architectures for Lifelong Learning of Everyday Human Behaviors AU - Lee, Sang-Woo AU - Lee, Chung-Yeon AU - Kwak, Dong-Hyun AU - Kim, Jiwon AU - Kim, Jeonghee AU - Zhang, Byoung-Tak T2 - Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-16 PB - AAAI Press / International Joint Conferences on Artificial Intelligence PY - 2016 SP - 1669 EP - 1674 UR - https://ijcai.org/proceedings/2016/239 ER -Copied
Machine-readable versions of this paper, for tools and AI agents: