Jaemin Yoo
Postdoctoral Researcher at Carnegie Mellon University
I am a postdoctoral research fellow at Carnegie Mellon University, working with Prof. Leman Akoglu. I received my Ph.D. and B.S. in Computer Science and Engineering from Seoul National University, where I was advised by Prof. U Kang. I am a recipient of the Google PhD Fellowship and the Qualcomm Innovation Fellowship. My research interests include graph mining, graph neural networks, time series forecasting, and interpretable tree models.
News
- [Aug. 2022] Our work on hypergraph reciprocity is accepted to ICDM 2022.
- [Aug. 2022] I served as a session chair at KDD 2022.
- [Aug. 2022] Our preprint on self-supervised anomaly detection is out [paper].
- [Aug. 2022] I gave an invited talk at SNU AI Summer School 2022.
- [Aug. 2022] I gave an invited talk at SK C&C.
- [Jul. 2022] I gave an invited talk at KAIST School of Computing [slides].
- [Jul. 2022] I gave an invited talk to AWS Deep Learning group.
- [May 2022] Our extended work on graph PU learning is accepted to KAIS.
- [May 2022] Our work on node feature estimation is accepted to KDD 2022 [paper].
- [Apr. 2022] I gave an invited talk at EIRIC Seminar [slides and video (Korean)].
- [Mar. 2022] I started my new career at Carnegie Mellon University.
- [Mar. 2022] Our work on signed random walks is accepted to PLOS ONE [paper].
- [Feb. 2022] I received the Best Ph.D. Thesis Award in SNU CSE.
- [Feb. 2022] I gave an invited talk at KAIST School of Electrical Engineering [slides].
- [Feb. 2022] I gave an invited talk at LG AI Research Tech Talk.
- [Jan. 2022] Our work on graph augmentation is accepted to WWW 2022 [paper].
- [Jan. 2022] Our work on hypergraph sampling is accepted to WWW 2022 [paper].
- [Dec. 2021] Our work on unifying tree models is accepted to SDM 2022 [paper].
- [Dec. 2021] Our work is selected as one of the best-ranked papers of ICDM 2021.
- [Oct. 2021] I gave an invited talk at KAIST AI Student Colloquium [slides].
About Me
Position and Education
-
Carnegie Mellon University
Postdoctoral Research Fellow (Mar. 2022 - Present)
Heinz College of Information Systems and Public Policy
DATA Lab (by Prof. Leman Akoglu) -
Seoul National University
Ph.D. in Computer Science and Engineering (Feb. 2022)
B.S. in Computer Science and Engineering (Feb. 2016)
Research Interests
- Missing feature estimation in graphs (KDD'22)
- Graph structure augmentation (WWW'22)
- Interpretable tree models (ICDM'19, PAKDD'21, SDM'22)
- Node classification in graphs (ICDM'17, IJCAI'19, ICDM'21)
- Multivariate time series forecasting (SDM'21, KDD'21)
- Subgraph sampling (WSDM'20)
- Zero-shot knowledge distillation (NeurIPS'19)
Awards and Honors
- Best Ph.D. Thesis Award in SNU CSE (Feb. 2022)
- One of the Best-Ranked Papers of ICDM 2021 (Dec. 2021)
- SIAM Student Travel Award (SDM 2021; Apr. 2021)
- Qualcomm Innovation Fellowship (Dec. 2020)
- Yulchon AI Star Award (Sep. 2020)
- Google PhD Fellowship (Machine Learning; Sep. 2019)
- Samsung HumanTech Paper Award (Feb. 2019)
- Google Conference Scholarship (Nov. 2017)
Publications
Preprints
-
Understanding the Effect of Data Augmentation in Self-supervised Anomaly Detection
Jaemin Yoo, Tiancheng Zhao, and Leman Akoglu
arXiv Preprint (2022) [ paper | bib ]
Tutorials
-
Mining of Real-world Hypergraphs: Concepts, Patterns, and Generators
Geon Lee, Jaemin Yoo, and Kijung Shin
ICDM 2022 / CIKM 2022 / DSAA 2022 [ slides ]
Conferences
-
Reciprocity in Directed Hypergraphs: Measures, Findings, and Generators
Sunwoo Kim, Minyoung Choe, Jaemin Yoo, and Kijung Shin
ICDM 2022 (to appear)
-
Accurate Node Feature Estimation with Structured Variational Graph Autoencoder
Jaemin Yoo, Hyunsik Jeon, Jinhong Jung, and U Kang
KDD 2022 [ paper | slides | code and datasets | bib ]
-
Model-Agnostic Augmentation for Accurate Graph Classification
Jaemin Yoo, Sooyeon Shim, and U Kang
WWW 2022 [ paper | slides | code and datasets | bib ]
-
MiDaS: Representative Sampling from Real-world Hypergraphs
Minyoung Choe, Jaemin Yoo, Geon Lee, Woonsung Baek, U Kang, and Kijung Shin
WWW 2022 [ paper | slides | code and datasets | bib ]
-
Transition Matrix Representation of Trees with Transposed Convolutions
Jaemin Yoo and Lee Sael
SDM 2022 [ paper | slides | code and datasets | bib ]
-
Accurate Graph-Based PU Learning without Class Prior
Jaemin Yoo*, Junghun Kim*, Hoyoung Yoon*, Geonsoo Kim, Changwon Jang, and U Kang (*equal contribution)
ICDM 2021 [ paper | slides | bib | blog (Korean) ]
One of the best-ranked papers of ICDM 2021 for fast-track journal invitation
-
Accurate Multivariate Stock Movement Prediction via Data-Axis Transformer with Multi-Level
Contexts
Jaemin Yoo, Yejun Soun, Yong-chan Park, and U Kang
KDD 2021 [ paper | slides | datasets | bib ]
-
Gaussian Soft Decision Trees for Interpretable Feature-Based Classification
Jaemin Yoo and Lee Sael
PAKDD 2021 [ paper | slides | code and datasets | bib ]
-
Attention-Based Autoregression for Accurate and Efficient Multivariate Time Series
Forecasting
Jaemin Yoo and U Kang
SDM 2021 [ paper | slides | bib | blog (Korean) ]
-
Sampling Subgraphs with Guaranteed Treewidth for Accurate and Efficient Graphical
Inference
Jaemin Yoo, U Kang, Mauro Scanagatta, Giorgio Corani, and Marco Zaffalon
WSDM 2020 [ paper | poster | code and datasets | bib | blog (Korean) ]
Samsung HumanTech Paper Award, Qualcomm Innovation Fellowship Korea
-
Knowledge Extraction with No Observable Data
Jaemin Yoo, Minyong Cho, Taebum Kim, and U Kang
NeurIPS 2019 [ paper | slides | code and datasets | poster | bib | blog (Korean) ]
Qualcomm Innovation Fellowship Korea
-
EDiT: Interpreting Ensemble Models via Compact Soft Decision Trees
Jaemin Yoo and Lee Sael
ICDM 2019 [ paper | slides | code and datasets | bib ]
-
Belief Propagation Network for Hard Inductive Semi-Supervised Learning
Jaemin Yoo, Hyunsik Jeon, and U Kang
IJCAI 2019 [ paper | slides | code and datasets | poster | bib | blog (Korean) ]
-
Fast and Scalable Distributed Loopy Belief Propagation on Real-World Graphs
Saehan Jo, Jaemin Yoo, and U Kang
WSDM 2018 [ paper | slides | poster | code and datasets | bib ]
-
Supervised Belief Propagation: Scalable Supervised Inference on Attributed Networks
Jaemin Yoo, Saehan Jo, and U Kang
ICDM 2017 [ paper | slides | code and datasets | bib | blog (Korean) ]
Regular paper (top 9.3% = 72/778)
Journals
-
Graph-based PU Learning for Binary and Multiclass Classification without Class Prior
Jaemin Yoo*, Junghun Kim*, Hoyoung Yoon*, Geonsoo Kim, Changwon Jang, and U Kang (*equal contribution)
Knowledge and Information Systems (2022) [ paper | bib ]
-
Signed Random Walk Diffusion for Effective Representation Learning in Signed Graphs
Jinhong Jung, Jaemin Yoo, and U Kang
PLOS ONE (2022) [ paper | code and datasets | bib ]
-
Efficient Learning of Bounded-Treewidth Bayesian Networks from Complete and Incomplete Data
Sets
Mauro Scanagatta, Giorgio Corani, Marco Zaffalon, Jaemin Yoo, and U Kang
International Journal of Approximate Reasoning (2018) [ paper | bib ]
Miscellaneous
Invited Talks
- SNU AI Summer School 2022 (Online, Aug. 2022)
- SK C&C (Online, Aug. 2022)
- KAIST School of Computing (Online, Jul. 2022)
- AWS Deep Learning Group (Online, Jul. 2022)
- EIRIC Seminar (Online, Apr. 2022)
- KAIST School of Electrical Engineering (Daejeon, Feb. 2022)
- LG AI Research Tech Talk (Seoul, Feb. 2022)
- KAIST AI Student Colloquium (Online, Oct. 2021)
- SNU AI Summer School 2021 (Online, Aug. 2021)
- SNU AI Institute (AIIS) Retreat 2021 (Seoul, Apr. 2021)
- NAVER Online Tech Talk (Online, Dec. 2020)
- SNU AI Summer School 2020 (Seoul, Aug. 2020)
- SNU Hospital (Seoul, Jul. 2020)
- SNU AI Institute (AIIS) Retreat 2020 (Seoul, Jun. 2020)
- Kakao Enterprise (Seongnam, Jan. 2020)
- Korea Software Congress 2019 (Pyeongchang, Dec. 2019)
- SNU Center for AI (SCAI) Retreat 2019 (Chuncheon, Jul. 2019)
- Samsung Electronics (Suwon, May 2019)
- IDSIA (Lugano, Jul. 2018)
- Korea Software Congress 2017 (Busan, Dec. 2017)
Jaemin Yoo @ CMU