Jaewon Chung

GPU cluster resource management

• Developing Crane, an elastic GPU cluster manager for deep learning workloads.

Distributed DNN inference, Mobile Computing

• Worked with Prof. Soo-Mook Moon on developing distributed DNN inference schemes.
• Proposed a distributed video DNN inference scheme that drastically reduces network traffic by exploiting intermittent knowledge distillation.
• Implemented the scheme with PyTorch and OpenMPI with an NVIDIA Jetson Nano embedded board as the client and conducted evaluation.

Meta-learning, Few-shot Classification

• Conducted research on better meta-initialization points for Model-Agnostic Meta-Learning (MAML) using an LSTM-based neural memory.
• Conducted research on generating task-aware class embeddings to augment feature maps of MAML by formulating a convex program (DPP).

Quantitative Susceptibility Mapping (QSM), Deep Learning

• Worked with Prof. Jongho Lee to submit our solution to the QSM challenge held by the 5th Int'l Workshop on MRI Phase Contrast and QSM.
• Designed, implemented, and trained a U-Net variant on in-vivo brain MRI field images and their COSMOS results.

A free research group on all domains of deep learning

• Gained experience extensively in computer vision and meta-learning, and attended talks on computer vision, natural language processing, reinforcement learning, and speech recognition.
• Gave a talk with the title "Memory plus Meta-Learning".

An official Coursera community that traslates Coursera lecture subtitles

• Served as Language Coordinator, a selected position that reviews and confirms works by other translators.
• Created Korean subtitles for Coursera Lectures initially provided only in English. Focused on courses related to machine learning.