On-device Efficient DNN Adaptation Framework for Real-time Applications
Apr 27, 2024·
,,·
1 min read

Hyeongheon Cha
Taesik Gong
Sung-Ju Lee

Position
Project leader in KAIST Network and Mobile System Lab
Project Goals & Works
- Aim to develop a novel time-efficient on-device DNN adaptation strategy for real-time mobile applications, balancing model accuracy and inference speed.
- Exploring model adaptation methods that enable real-time inference on extremely resource-constrained devices.