miscellaneous-1

imbalanced classification

3260 papers • 126 benchmarks • 313 datasets

learning classifier from class-imbalanced data

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in imbalanced-classification-1

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find imbalanced-classification-1 models and implementations

Datasets

Kinetics

CI-MNIST

WikiChurches

Industrial Benchmark Dataset for Customer Escalation Prediction

kaggle stroke Prediction competition

Subtasks

No subtasks available.

Most implemented papers

MoleculeNet: a benchmark for molecular machine learning

Bharath Ramsundar, Zhenqin Wu, Evan N. Feinberg, Joseph Gomes, Caleb Geniesse, Aneesh S. Pappu, K. Leswing, V. Pande•Wed Mar 01 2017

A large scale benchmark for molecular machine learning consisting of multiple public datasets, metrics, featurizations and learning algorithms.

2218

Content

Imbalanced-MiniKinetics200

Can you predict product backorder?

0

Paper Graph

SoftMatch: Addressing the Quantity-Quality Trade-off in Semi-supervised Learning

Jindong Wang, B. Raj, B. Schiele, M. Savvides, Yidong Wang, R. Tao, Hao Chen, Yue Fan, Xingxu Xie•Wed Jan 25 2023

This paper revisits the popular pseudo-labeling methods via a unified sample weighting formulation and demonstrates the inherent quantity-quality trade-off problem of pseudo-labels with thresholding, which may prohibit learning.

215 0

Paper Graph

Deep reinforcement learning for imbalanced classification

Enlu Lin, Qiong Chen, Xiaoming Qi•Fri Jan 04 2019

A general imbalanced classification model based on deep reinforcement learning, in which the problem is formulated as a sequential decision-making process and solved by a deep Q-learning network, and the agent finally finds an optimal classification policy in imbalanced data.

200 0

Paper Graph

Large-Scale Long-Tailed Recognition in an Open World

Ziwei Liu, Boqing Gong, Zhongqi Miao, Xiaohang Zhan, Jiayun Wang, Stella X. Yu•Tue Apr 09 2019

An integrated OLTR algorithm is developed that maps an image to a feature space such that visual concepts can easily relate to each other based on a learned metric that respects the closed-world classification while acknowledging the novelty of the open world.

1330 0

Paper Graph

MNIST-MIX: a multi-language handwritten digit recognition dataset

Weiwei Jiang•Tue Apr 07 2020

A multi-language handwritten digit recognition dataset named MNIST-MIX, which is the largest dataset of the same type in terms of both languages and data samples, and a LeNet model which is pre-trained on MNIST as the baseline is presented.

27 0

Paper Graph

Long-tailed Recognition by Routing Diverse Distribution-Aware Experts

Ziwei Liu, Zhongqi Miao, Stella X. Yu, Xudong Wang, Long Lian•Sun Oct 04 2020

RIDE aims to reduce both the bias and the variance of a long-tailed classifier by RoutIng Diverse Experts (RIDE), which significantly outperforms the state-of-the-art methods by 5% to 7% on all the benchmarks including CIFAR100-LT, ImageNet-LT and iNaturalist.

462 0

Paper Graph

MESA: Boost Ensemble Imbalanced Learning with MEta-SAmpler

Jiang Bian, Wei Cao, Jing Jiang, Yi Chang, Zhining Liu, Pengfei Wei•Fri Oct 16 2020

This paper introduces a novel ensemble IL framework named MESA, which adaptively resamples the training set in iterations to get multiple classifiers and forms a cascade ensemble model that makes MESA generally applicable to most of the existing learning models and the meta-sampler can be efficiently applied to new tasks.

74 0

Paper Graph

A Large-Scale Database for Graph Representation Learning

Yuxiao Dong, Duen Horng Chau, Scott Freitas, Joshua Neil•Sun Nov 15 2020

This work introduces Mal net, the largest public graph database ever constructed, representing a large-scale ontology of software function call graphs, and provides a detailed analysis of MalNet, discussing its properties and provenance.

75 0

Paper Graph

Box drawings for learning with imbalanced data

C. Rudin, Siong Thye Goh•Wed Mar 12 2014

This work proposes two machine learning algorithms to handle highly imbalanced classification problems, one of which follows a \textit{characterize then discriminate} approach, where the positive class is characterized first by boxes, and then each box boundary becomes a separate discriminative classifier.

43 0

Paper Graph

Boosting with Lexicographic Programming: Addressing Class Imbalance without Cost Tuning

Shounak Datta, Sayak Nag, Swagatam Das•Wed Aug 30 2017

It is shown that the assignment of weights to the component classifiers of a boosted ensemble can be thought of as a game of Tug of War between the classes in the margin space, and demonstrated how this insight can be used to attain a good compromise between the rare and abundant classes without having to resort to cost set tuning.

32 0

Paper Graph

Adding a benchmark result helps the community track progress.