AI/machine learning efforts are already a big part of ongoing research in all 6 divisions –
Computer, Communication, Signal, Wave, Circuit and Device – of KAIST EE. Examples include
neuromorphic devices, VLSI hardware architecture tailored to machine learning, image/voice recognition
via deep learning, statistical inference, coding and information theory to enhance distributed machine
learning, intelligent robots, quantum information, brain imaging, etc.
In recent years, artificial neural networks (ANN) including deep neural network (DNN) and spiking neural network (SNN) have achieved unprecedented accuracies in large-scale recognition and classification tasks by utilizing supercomputing resources. While several application-specific integrated circuit (ASIC) solutions utilizing conventional CMOS devices have been previously proposed, limitations still exist on energy consumptions, online learning capabilities and chip density. To address all issues in AI hardware, the community is moving towards utilizing emerging AI devices as artificial neurons and synapses because they can offer fast parallel computing at extremely small device footprint with low power consumption. The main research interest of Device Division of KAIST EE is to develop large-scale neural network arrays for artificial intelligence (AI) hardware based on new design of artificial neurons and synapses.
See below for specific ongoing research topics related to AI and machine learning within the Device Division of KAIST EE.