no code implementations • 16 Feb 2024 • Hyeonsu Jeong, Hye Won Chung
By deriving a closed-form solution for the student model's outputs, we discover that SD essentially functions as label averaging among instances with high feature correlations.
1 code implementation • 31 May 2023 • Joonhyuk Yang, Dongpil Shin, Hye Won Chung
We consider the problem of graph matching, or learning vertex correspondence, between two correlated stochastic block models (SBMs).
no code implementations • 12 Jan 2023 • Ji Hyung Jung, Hye Won Chung, Ji Oon Lee
We first show that the principal component analysis can be improved by entrywise pre-transforming the data matrix if the noise is non-Gaussian, generalizing the known results for the spiked random matrix models with rank-1 signals.
1 code implementation • 3 Jan 2023 • Nohyun Ki, Hoyong Choi, Hye Won Chung
Many recent works on understanding deep learning try to quantify how much individual data instances influence the optimization and generalization of a model.
no code implementations • 29 Dec 2022 • Hyeonsu Jeong, Hye Won Chung
Under this model, we propose a two-stage inference algorithm to infer both the top two answers and the confusion probability.
2 code implementations • 8 Jul 2022 • Minguk Jang, Sae-Young Chung, Hye Won Chung
To overcome this limitation, we propose a novel test-time adaptation method, called Test-time Adaptation via Self-Training with nearest neighbor information (TAST), which is composed of the following procedures: (1) adds trainable adaptation modules on top of the trained feature extractor; (2) newly defines a pseudo-label distribution for the test data by using the nearest neighbor information; (3) trains these modules only a few times during test time to match the nearest neighbor-based pseudo label distribution and a prototype-based class distribution for the test data; and (4) predicts the label of test data using the average predicted class distribution from these modules.
no code implementations • 2 Mar 2022 • Hye Won Chung, Jiho Lee, Ji Oon Lee
For general non-Gaussian noise, assuming that the signal is drawn from the Rademacher prior, we prove that the log likelihood ratio (LR) of the spiked model against the null model converges to a Gaussian when the signal-to-noise ratio is below a certain threshold.
1 code implementation • 19 Nov 2021 • Doyeon Kim, Jeonghwan Lee, Hye Won Chung
Inferring correct labels from multiple noisy answers on data, however, has been a challenging problem, since the quality of the answers varies widely across tasks and workers.
no code implementations • 28 Apr 2021 • Ji Hyung Jung, Hye Won Chung, Ji Oon Lee
We show that the principal component analysis can be improved by pre-transforming the matrix entries if the noise is non-Gaussian.
1 code implementation • NeurIPS 2021 • Jinhee Lee, HaeRi Kim, Youngkyu Hong, Hye Won Chung
To promote diversity in sample generation without degrading the overall quality, we propose a simple yet effective method to diagnose and emphasize underrepresented samples during training of a GAN.
no code implementations • 23 Mar 2020 • Jeonghwan Lee, Daesung Kim, Hye Won Chung
We study hypergraph clustering in the weighted $d$-uniform hypergraph stochastic block model ($d$\textsf{-WHSBM}), where each edge consisting of $d$ nodes from the same community has higher expected weight than the edges consisting of nodes from different communities.
no code implementations • 21 Mar 2020 • Do-Yeon Kim, Hye Won Chung
We consider crowdsourced labeling under a $d$-type worker-task specialization model, where each worker and task is associated with one particular type among a finite set of types and a worker provides a more reliable answer to tasks of the matched type than to tasks of unmatched types.
no code implementations • 31 Jan 2020 • Daesung Kim, Hye Won Chung
In particular, we consider the problem of classifying $m$ binary labels with XOR queries that ask whether the number of objects having a given attribute in the chosen subset of size $d$ is even or odd.
no code implementations • 16 Jan 2020 • Ji Hyung Jung, Hye Won Chung, Ji Oon Lee
We study the statistical decision process of detecting the signal from a `signal+noise' type matrix model with an additive Wigner noise.
no code implementations • 19 Apr 2019 • Youngjae Min, Hye Won Chung
This paper constructs shallow sigmoid-type neural networks that achieve 100% accuracy in classification for datasets following a linear separability condition.
no code implementations • 28 Sep 2018 • Hye Won Chung, Ji Oon Lee
We propose a hypothesis test on the presence of the signal by utilizing the linear spectral statistics of the data matrix.
no code implementations • 4 Sep 2018 • Hye Won Chung, Ji Oon Lee, Do-Yeon Kim, Alfred O. Hero
We define the query difficulty $\bar{d}$ as the average size of the query subsets and the sample complexity $n$ as the minimum number of measurements required to attain a given recovery accuracy.