Classification with noisy labels by importance reweighting

T Liu, D Tao - IEEE Transactions on pattern analysis and …, 2015 - ieeexplore.ieee.org
In this paper, we study a classification problem in which sample labels are randomly
corrupted. In this scenario, there is an unobservable sample with noise-free labels …

[KSIĄŻKA][B] Density ratio estimation in machine learning

M Sugiyama, T Suzuki, T Kanamori - 2012 - books.google.com
Machine learning is an interdisciplinary field of science and engineering that studies
mathematical theories and practical applications of systems that learn. This book introduces …

Change-point detection in time-series data by relative density-ratio estimation

S Liu, M Yamada, N Collier, M Sugiyama - Neural Networks, 2013 - Elsevier
The objective of change-point detection is to discover abrupt property changes lying behind
time-series data. In this paper, we present a novel statistical change-point detection …

[KSIĄŻKA][B] Introduction to statistical machine learning

M Sugiyama - 2015 - books.google.com
Machine learning allows computers to learn and discern patterns without actually being
programmed. When Statistical techniques and machine learning are combined together they …

Few-shot domain adaptation by causal mechanism transfer

T Teshima, I Sato, M Sugiyama - … Conference on Machine …, 2020 - proceedings.mlr.press
We study few-shot supervised domain adaptation (DA) for regression problems, where only
a few labeled target domain data and many labeled source domain data are available. Many …

Maximum mean discrepancy test is aware of adversarial attacks

R Gao, F Liu, J Zhang, B Han, T Liu… - International …, 2021 - proceedings.mlr.press
The maximum mean discrepancy (MMD) test could in principle detect any distributional
discrepancy between two datasets. However, it has been shown that the MMD test is …

Alleviating semantics distortion in unsupervised low-level image-to-image translation via structure consistency constraint

J Guo, J Li, H Fu, M Gong… - Proceedings of the …, 2022 - openaccess.thecvf.com
Unsupervised image-to-image (I2I) translation aims to learn a domain map** function that
can preserve the semantics of the input images without paired data. However, because the …

Estimating and explaining model performance when both covariates and labels shift

L Chen, M Zaharia, JY Zou - Advances in Neural …, 2022 - proceedings.neurips.cc
Deployed machine learning (ML) models often encounter new user data that differs from
their training data. Therefore, estimating how well a given model might perform on the new …

Maximum spatial perturbation consistency for unpaired image-to-image translation

Y Xu, S **e, W Wu, K Zhang, M Gong… - Proceedings of the …, 2022 - openaccess.thecvf.com
Unpaired image-to-image translation (I2I) is an ill-posed problem, as an infinite number of
translation functions can map the source domain distribution to the target distribution …

Adapting to continuous covariate shift via online density ratio estimation

YJ Zhang, ZY Zhang, P Zhao… - Advances in Neural …, 2024 - proceedings.neurips.cc
Dealing with distribution shifts is one of the central challenges for modern machine learning.
One fundamental situation is the covariate shift, where the input distributions of data change …