Instance-Dependent Label-Noise Learning with Manifold-Regularized Transition Matrix Estimation

by   De Cheng, et al.
Chongqing University of Post and Telecommunications
The University of Sydney
Hong Kong Baptist University
Xidian University
The University of Tokyo

In label-noise learning, estimating the transition matrix has attracted more and more attention as the matrix plays an important role in building statistically consistent classifiers. However, it is very challenging to estimate the transition matrix T(x), where x denotes the instance, because it is unidentifiable under the instance-dependent noise(IDN). To address this problem, we have noticed that, there are psychological and physiological evidences showing that we humans are more likely to annotate instances of similar appearances to the same classes, and thus poor-quality or ambiguous instances of similar appearances are easier to be mislabeled to the correlated or same noisy classes. Therefore, we propose assumption on the geometry of T(x) that "the closer two instances are, the more similar their corresponding transition matrices should be". More specifically, we formulate above assumption into the manifold embedding, to effectively reduce the degree of freedom of T(x) and make it stably estimable in practice. The proposed manifold-regularized technique works by directly reducing the estimation error without hurting the approximation error about the estimation problem of T(x). Experimental evaluations on four synthetic and two real-world datasets demonstrate that our method is superior to state-of-the-art approaches for label-noise learning under the challenging IDN.


Parts-dependent Label Noise: Towards Instance-dependent Label Noise

Learning with the instance-dependent label noise is challenging, because...

Approximating Instance-Dependent Noise via Instance-Confidence Embedding

Label noise in multiclass classification is a major obstacle to the depl...

Are Anchor Points Really Indispensable in Label-Noise Learning?

In label-noise learning, noise transition matrix, denoting the probabili...

Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learning

The transition matrix, denoting the transition relationship from clean l...

Transferring Annotator- and Instance-dependent Transition Matrix for Learning from Crowds

Learning from crowds describes that the annotations of training data are...

Provably End-to-end Label-Noise Learning without Anchor Points

In label-noise learning, the transition matrix plays a key role in build...

Inaccurate Label Distribution Learning

Label distribution learning (LDL) trains a model to predict the relevanc...

Please sign up or login with your details

Forgot password? Click here to reset