Machine Learning Lab
Jungseul Ok, Dongwoo Kim (Computer Science and Engineering)
An excellent teacher knows how to help students easily understand even the most difficult subject. Using their unique skills, they improve the learning outcomes of students. Artificial intelligence is much the same in that it also needs to be taught by such top-notch “teachers” in learning data to mimic human intelligence.
The Machine Learning Lab directed by professors Jungseul Ok and Dongwoo Kim at the Department of Computer Science and Engineering, POSTECH, probes into machine learning theories and their applications that form the groundwork of artificial intelligence. Its primary research theme is to develop efficient machine learning algorithms to engineer AI capable of generating optimal outcomes with minimal data.
Another recent research project concerns the methodology to enable AI to learn unstructured data. Unlike structured data, which is organized in an orderly manner in a predefined form, unstructured data exists randomly at any location and includes texts, audio, and videos. The Lab has created an algorithm to help AI learn disorderly-arranged data by presenting such data through graphs.
Furthermore, the Lab theoretically demonstrated how difficult it is to trace the transmission route of COVID-19 in the presence of infected individuals with unknown transmission routes, and then engineered AI to identify the transmission routes.
Although the Machine Learning Lab is still in its infancy, it has set an ambitious goal of developing efficient machine learning algorithms and turning AI into ‘public goods’ that are easily accessible by anyone. While research labs are normally operated by a single professor, the Lab is jointly headed by two professors. This means that students can receive instructions from both professors and actively share in one other’s research information to easily establish the appropriate research direction forward.
Professors Ok and Kim are joining hands not just to teach knowledge to AI, but also to find ways to better guide their students at the Lab.
Head of Lab
Institute of Artificial Intelligence 233