A Gaussian Process-Bayesian Bernoulli Mixture Model for Multi-Label Active Learning

no code implementations • NeurIPS 2021 • Weishi Shi, Dayou Yu, Qi Yu

However, data annotation for training MLC models becomes much more labor-intensive due to the correlated (hence non-exclusive) labels and a potential large and sparse label space.

Active Learning Inductive Bias +2

Uncertainty-Aware Multiple Instance Learning from Large-Scale Long Time Series Data

no code implementations • 16 Nov 2021 • Yuansheng Zhu, Weishi Shi, Deep Shankar Pandey, Yang Liu, Xiaofan Que, Daniel E. Krutz, Qi Yu

We propose a novel framework to classify large-scale time series data with long duration.

Multiple Instance Learning Time Series +1

Multifaceted Uncertainty Estimation for Label-Efficient Deep Learning

no code implementations • NeurIPS 2020 • Weishi Shi, Xujiang Zhao, Feng Chen, Qi Yu

We present a novel multi-source uncertainty prediction approach that enables deep learning (DL) models to be actively trained with much less labeled data.

Integrating Bayesian and Discriminative Sparse Kernel Machines for Multi-class Active Learning

no code implementations • NeurIPS 2019 • Weishi Shi, Qi Yu

We propose a novel active learning (AL) model that integrates Bayesian and discriminative kernel machines for fast and accurate multi-class data sampling.

Active Learning

Evidence-Aware Entropy Decomposition For Active Deep Learning

no code implementations • 25 Sep 2019 • Weishi Shi, Xujiang Zhao, Feng Chen, Qi Yu

We present a novel multi-source uncertainty prediction approach that enables deep learning (DL) models to be actively trained with much less labeled data.

Density Estimation