paper-with-me

Papers

Complementary-Label Learning for Arbitrary Losses and Models

2018-10-10 · Proceedings of the 36th International Conference on Machine Learning, 2019 6 · Takashi Ishida, Gang Niu, Aditya Krishna Menon, Masashi Sugiyama

In contrast to the standard classification paradigm where the true class is given to each training pattern, complementary-label learning only uses training patterns each equipped with a complementary label, which only specifies one of the classes that the pattern does not belong to. The goal of this paper is to derive a novel framework of complementary-label learning with an unbiased estimator of the classification risk, for arbitrary losses and models---all existing methods have failed to achieve this goal. Not only is this beneficial for the learning stage, it also makes model/hyper-parameter selection (through cross-validation) possible without the need of any ordinarily labeled validation data, while using any linear/non-linear models or convex/non-convex loss functions. We further improve the risk estimator by a non-negative correction and gradient ascent trick, and demonstrate its superiority through experiments.

📄 PDF Abstract BibTeX arXiv:1810.04327

Code (1)

takashiishida/comp 공식 구현 pytorch

Tasks

General ClassificationImage Classification

Similar Papers 제목 키워드 기반

Multi-Complementary and Unlabeled Learning for Arbitrary Losses and Models

2020-01-13 · Yuzhou Cao, Shuqi Liu, Yitian Xu

A weakly-supervised learning framework named as complementary-label learning has been proposed recently, where each sample is equipped with a single complementary label that denotes one of the classes the sample does not…

General ClassificationImage ClassificationWeakly-supervised Learning

SCL: Towards Accurate Domain Adaptive Object Detection via Gradient Detach Based Stacked Complementary Losses

2019-11-06 · Zhiqiang Shen, Harsh Maheshwari, Weichen Yao, Marios Savvides

Unsupervised domain adaptive object detection aims to learn a robust detector in the domain shift circumstance, where the training (source) domain is label-rich with bounding box annotations, while the testing (target) d…

object-detectionObject Detection

A Universal Unbiased Method for Classification from Aggregate Observations

2023-06-20 · Zixi Wei, Lei Feng, Bo Han, Tongliang Liu 외

In conventional supervised classification, true labels are required for individual instances. However, it could be prohibitive to collect the true labels for individual instances, due to privacy concerns or unaffordable …

ClassificationMultiple Instance Learning

Surrogate Losses in Passive and Active Learning

2012-07-16 · Steve Hanneke, Liu Yang

Active learning is a type of sequential design for supervised machine learning, in which the learning algorithm sequentially requests the labels of selected instances from a large pool of unlabeled data points. The objec…

Active Learning

Reduction from Complementary-Label Learning to Probability Estimates

2022-09-20 · Wei-I Lin, Hsuan-Tien Lin

Complementary-Label Learning (CLL) is a weakly-supervised learning problem that aims to learn a multi-class classifier from only complementary labels, which indicate a class to which an instance does not belong. Existing…

Weakly-supervised Learning