On June 19, KAIST AI/ML and Clova AI held the second joint AI workshop. Leading professors and veteran researchers in artificial intelligence and machine learning gave lectures about their recent publications.
Lectures from the workshop (in Korean)
14:00-14:10 | Opening Remarks and AI in KAIST | Prof. Alice Oh |
14:10-14:30 | ML for Underwater Perception | Prof. Ayoung Kim |
14:30-14:50 | Human-AI Interaction | Prof. Juho Kim |
14:50-15:10 | Lightweight Visual Recognition | Youngjoon Yoo Ph.D. |
15:10-15:30 | Large-Scale Graph Mining | Prof. Kijung Shin |
15:30-16:30 | Poster Session and Coffee Break | KAIST AI Labs |
16:30-16:50 | Latest ML Research | Prof. Jinwoo Shin |
16:50-17:10 | Information Theory-Based ML | Prof. Hye Won Chung |
17:10-17:30 | Curiosity-Based RL and Practical AutoML | Jihoon Kim Ph.D. |
17:30-17:50 | Computational Social Science Problems | Prof. Meeyoung Cha |
17:50-18:00 | Closing Remarks and Research Opportunity in Clova AI | Prof. Sung Kim |





Between the lectures, 24 students from KAIST AI/ML also delivered a poster presentation on their research works.


Name | Research Advisor | Poster Title | The Venue where the Research was Originally Presented (Conference, Journal, or under Review) |
Hwanjun Song | Prof. Jae-Gil Lee | SELFIE: Refurbishing Unclean Samples for Robust Deep Learning | ICML 2019 |
송경우 | Prof. Il-Chul Moon | Hierarchical Context enabled Recurrent Neural Network for Recommendation | AAAI-19 |
Kyunghwan Son | Prof. Yung Yi | QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement learning | ICML 2019 |
Dae Woo Kim | Prof. Yung Yi | Learning to Schedule Communication in Multi-agent Reinforcement Learning | ICLR 2019 |
Wissam J. Baddar | Prof. Yong Man Ro | Mode Variational LSTM Robust to Unseen Modes of Variation: Application to Facial Expression Recognition | AAAI |
Hyungseok Song | Prof. Yung Yi | Solving Continual Combinatorial Selection via Deep Reinforcement Learning | IJCAI |
Raehyuk Jung | Prof. Jean-Charles Bazin | Deep360Up: Deep learning-based approach for automatic upright adjustment of VR images | Published in IEEE VR 2019 |
Taewon Jeong | Prof. Heeyoung Kim | Ladder Capsule Network | ICML 2019 |
Dongkwan Kim | Prof. Alice Oh | Homogeneity-Based Transmissive Process To Model True and False News in Social Networks | WSDM 2019 |
JinYeong Bak | Prof. Alice Oh | Variational Hierarchical User-based Conversation Model | Under Review |
박성준 | Prof. Alice Oh | Conversation model fine tuning for classifying client’s utterances in counseling dialogues | NAACL 2019 |
Giwon Hong | Prof. Sung-Hyon Myaeng | Aligning Open IE Relations and KB Relations using a Siamese Network Based on Word Embedding | IWCS 2019 |
Taehyun Kim | Prof. Se-Young Yun | Orthogonal Feature Regularization: A Novel Approach for training robust models | Under review of NeurIPS 2019 |
Taegyun Kwon | Prof. Juhan Nam | Graph Neural Network for Music Score Dataand Modeling Expressive Piano Performance | ICML 2019 |
김수민 | Prof. Sungeui Yoon | Single Image Reflection Removal with Physically-based Rendering | Under review of ICCV |
Tae Kihyun | Prof. Steven Euijong Whang | Data Cleaning for Accurate, Fair, and Robust Models: A Big Data – AI Integration Approach | 3rd Int’l Workshop on Data Management for End-to-End Machine Learning, DEEM @ ACM SIGMOD, June 2019 |
Hankook Lee | Prof. Jinwoo Shin | Learning What and Where to Transfer | ICML2019 |
Jongheon Jeong | Prof. Jinwoo Shin | Training CNNs with Selective Allocation of Channels | ICML 2019 |
Geon Heo | Prof. Steven Whang | Inspector Gadget: Massive Image Labeling for Machine Learning | under review |
Jinhan Kim | Prof. Shin Yoo | Guiding Deep Learning System Testing using Surprise Adequacy | ICSE 2019 |
Giseop Kim | Prof. Ayoung Kim | 1-Day Learning, 1-Year Localization: Long-term LiDAR Localization using Scan Context Image | ICRA 2019 |
Hajin Shim | Prof. Eunho Yang | Joint Active Feature Acquisition and Classification with Variable-Size Set Encoding | NeurIPS 2018 |
Jinwon Yoon | Prof. Hwasoo Yeo | AI-based Mobility: Traffic Predictive Solutions for Large Urban Area | Conference |
Among 24 presenters, Jiyoung Han and Taegyun Kwon won the best post award and received the cash prize of 500,000 South Korean Won each.

(From left to right) Alice Oh, Jiyoung Han, and Sung Kim

(From left to right) Sung Kim, Taegyun Kwon, and Alice Oh