A resource-constrained edge computer (e.g., smartphone, coral dev board) used to obtain local information via its sensors, perform simple pre-processing, and deliver the information to the cloud for more serious analysis. However, recent development of lightening techniques for deep neural networks (DNN) and low-cost AI accelerators, such as edge TPU and mobile GPU, results in a new paradigm "Ambient AI." Under Ambient AI paradigm, the edge computer does not have to rely on the cloud anymore but performs "AI computing" locally (e.g., inference through a DNN model by itself). This new paradigm has the potential to enable a variety of AI applications by improving latency, privacy, and network bandwidth.

This course introduces important techniques which enable this paradigm (including those provided by TensorFlow Lite). Students will learn these techniques both through lectures and hands-on experiences. To this end, this course guides students to get familiar with TensorFlow and TensorFlow Lite, teaching them how to implement various DNNs on TensorFlow, from MLP and CNN to object detection frameworks. Then students will run TensorFlow and TensorFlow Lite on not only their own computers but also Ambient AI platforms (just smaller Linux computers), such as Coral dev boards and Jetson Nano. In terms of application, this course will introduce computer vision but project topics do not have to be limited to this. 

Here are some projects demos in the previous semesters

This course, as an introductory course for Graduate School of Data Science (GSDS), provides basic knowledge for students lacking computer background so that they can follow advanced computing courses at GSDS and do relative research. This course lets students study basic principles and concepts of programming, how to efficiently organize and manipulate data,  and how a program is executed by the underlying computer systems . In doing so, most of the class will be based on Python language due to its convenience and wide usage. At the same time, students will do experience and program in C language a bit, see that Python is more convenient but C is faster and safe, and understand the reason why.  

Awards & Grants

• Best Practice for Untact Education, Faculty of Liberal Education at Seoul National University, 2021.

• Modularization Planning and Development of Programming Foundation Course, 빅데이터 혁신공유대학, 2021.10 ~ 2022.01

• Development of Hybrid Large Class Model, Faculty of Liberal Education at Seoul National University, 2021.09 ~ 2021.12