News
2024
(Aug. 2024) PI for Basic Research Lab (BRL) on the topic of Theoretical Framework for Foundation Models
Jointly with Prof. Jaekyun Moon (KAIST), Prof. Jy-yong Sohn (Yonsei), and Prof. Dohyun Kwon (Univ. of Seoul), our group will develop a theoretical framework and algorithms for foundation models. Great thanks to National Research Foundation (NRF).
(Aug. 2024) A new paper is accepted to TMLR
Representation Norm Amplification for Out-of-Distribution Detection in Long-Tail Learning [paper]
(July 2024) Jointly with Lele Wang (UBC), I delivered a tutorial at ISIT 2024 on the topic of "Graph Matching: Fundamental Limits and Efficient Algorithms." Here are the slides (Part I, Part II).
(Mar. 2024) A new paper is accepted to IEEE TIT
Detection Problems in the Spiked Random Matrix Models [arXiv]
(Apr. 2024) Two papers are accepted to ICML 2024
BWS: Best Window Selection Based on Sample Scores for Data Pruning across Broad Ranges
SelMatch: Effectively Scaling Up Dataset Distillation via Selection-Based Initialization and Partial Updates by Trajectory Matching
(Apr. 2024) I will give a tutorial at IEEE ISIT 2024 on the topic of "Graph Matching: Fundamental Limits and Efficient Algorithms"
(Apr. 2024) A new paper is accepted to IEEE ISIT 2024
Exact Graph Matching in Correlated Gaussian-Attributed Erods-Renyi Model
(Mar. 2024) A new paper is published in IEEE TIT
A Worker-Task Specialization Model for Crowdsourcing: Efficient Inference and Fundamental Limits [arXiv]
(Feb. 2024) Delivered an invited talk at UCSD ITA on the topic of understanding self-distillation in multi-class classification
(Jan. 2024) I will be serving as an organizing committee of ISIT 2024
2023
(Sep. 2023) A new paper is accepted to NeurIPS 2023
Rank-1 Matrix Completion with Gradient Descent and Small Random Initialization [arXiv]
(Sep. 2023) A paper is accepted to IEEE Trans. on Information Theory
A Worker-Task Specialization Model for Crowdsourcing: Efficient Inference and Fundamental Limits [arXiv]
(Aug. 2023) A paper is accepted to Allerton 2023
Graph Matching in Correlated Stochastic Block Models for Improved Graph Clustering
(Apr. 2023) Two papers are accepted to ICML2023
Recovering Top-Two Answers and Confusion Probability in Multi-Choice Crowdsourcing [arXiv]
Efficient Algorithms for Exact Graph Matching on Correlated Stochastic Block Models with Constant Correlation
(Mar. 2023) Received Departmental Outstanding Teaching Award for EE623 Information Theory.
(Feb. 2023) Delivered an invited talk at UCSD ITA on the topic of "Data Valuation without Training of a Model".
(Feb. 2023) Two first PhD graduates from our lab, Daesung Kim and Doyeon Kim. Congrats!
(Feb. 2023) Daesung has been awarded College of Engineering PhD Dissertation award. Congrats!
(Jan. 2023) Two papers are accepted to ICLR2023
Data Valuation without Training of a Model [arXiv]
Test-Time Adaptation via Self-Training with Nearest Neighbor Information [paper]
(Jan. 2023) Two new papers are available at arXiv.
2022
(Dec. 2022) A new paper is available at arXiv.
Rank-1 Matrix Completion with Gradient Descent and Small Random Initialization [arXiv]
(Nov. 2022) A new paper is published at IEEE Transactions on Information Theory.
Weak Detection in the Spiked Wigner Model [link]
(Oct. 2022) Delivered an invited talk at Institute of Data Science at National University of Singapore.
Data Valuation: Understanding Value of Data in Training Neural Networks [talk]
(Apr. 2022) Delivered an invited talk at "Post-Shannon Communication" panel session at JCCI 2022.
(Apr. 2022) A paper accepted in ISIT 2022.
A Generalized Worker-Task Specialization Model for Crowdsourcing: Optimal Limits and Algorithm [arXiv]
(Mar. 2022) An invited talk at Conference on Information Sciences and Systems (CISS) at Princeton
(Mar. 2022) Daesung has been awarded KAIST EE Best Research Achievement Award. Congrats Daesung!
(Mar. 2022) Two new papers are available at arXiv
A Worker-Task Specialization Model for Crowdsourcing: Efficient Inference and Fundamental Limits [arXiv]
Asymptotic Normality of Log-Likelihood Ratio and Fundamental Limit of the Weak Detection for Spiked Wigner Matrices [arXiv]
(Mar. 2022) Welcome new members, Youngmin Lee and Seunghun Cha
2021
(Dec. 2021) I received KAIST Technology Innovation Award.
(Sep. 2021) Our work on Self-Diagnosing GAN is accepted to NeurIPS 2021.
Self-Diagnosing GAN: Diagnosing Underrepresented Samples in Generative Adversarial Networks [arXiv]
(May 2021) Our work on detection of signal in the spiked rectangular models is accepted to ICML 2021.
Detection of Signal in the Spiked Rectangular Models [arXiv]
(May 2021) Our work on crowdsourced labelling is accepted to ISIT 2021.
Crowdsourced Labelling for Worker-Task Specialization Model [arXiv]
(Apr. 2021) Our work on binary classification with XOR queries is accepted to IEEE Transactions on Information Theory.
Binary Classification with XOR Queries: Fundamental Limits and An Efficient Algorithm [arXiv]
(Apr. 2021) A talk at ETRI-AI Academy (2nd) with the title "Data Optimization for AI."
(Mar. 2021) Supported by NRF Excellent Early-Career Research Funding for the next 4 years.
(Feb. 2021) A tutorial talk at KICS winter conference with the title "Data Optimization for AI."
2020
(Dec. 2020) A talk at ETRI-AI Academy (1st) with the title "Data Optimization for AI."
(Nov. 2020) Our work on hypergraph clustering is accepted to IEEE Journal on Selected Areas in Information Theory (JSAIT).
Robust Hypergraph Clustering via Convex Relaxation of Truncated MLE [arXiv]
(Apr. 2020) Our work on XOR queries for crowdsourced classification is accepted to ISIT 2020.
Crowdsourced Classification with XOR Queries: Fundamental Limits and An Efficient Algorithm [arXiv]
(Mar. 2020) A new paper is available at arXiv.
Hypergraph Clustering in the Weighted Stochastic Block Model via Convex Relaxation of Truncated MLE [arXiv]
(Feb. 2020) A new paper is available at arXiv.
Crowdsourced Classification with XOR Queries: Fundamental Limits and An Efficient Algorithm [arXiv]
(Jan. 2020) A new paper is available at arXiv.
Weak Detection of Signal in the Spiked Wigner Model with General Rank [arXiv]
2019
(Apr. 2019) Our work on weak detection of signal in the spiked Wigner model is accepted to ICML 2019.
Weak Detection of Signal in the Spiked Wigner Model [PMLR]
(Mar. 2019) Our work on structure design of neural network is accepted to ISIT 2019.
Shallow Neural Network can Perfectly Classify an Object following Separable Probability Distribution [arXiv]
2018
(Oct. 2018) Our work on coded crowdsourcing was presented at ISIT 2018 and Allerton 2018.
Parity Crowdsourcing for Cooperative Labeling [arXiv]