Lung and pancreatic tumor characterization in the deep learning era: novel supervised and unsupervised learning approaches

S Hussein, P Kandel, CW Bolan… - IEEE transactions on …, 2019 - ieeexplore.ieee.org
Risk stratification (characterization) of tumors from radiology images can be more accurate
and faster with computer-aided diagnosis (CAD) tools. Tumor characterization through such …

Weldon: Weakly supervised learning of deep convolutional neural networks

T Durand, N Thome, M Cord - Proceedings of the IEEE …, 2016 - openaccess.thecvf.com
In this paper, we introduce a novel framework for WEakly supervised Learning of Deep
cOnvolutional neural Networks (WELDON). Our method is dedicated to automatically …

Less is more: Learning highlight detection from video duration

B Xiong, Y Kalantidis, D Ghadiyaram… - Proceedings of the …, 2019 - openaccess.thecvf.com
Highlight detection has the potential to significantly ease video browsing, but existing
methods often suffer from expensive supervision requirements, where human viewers must …

Ensemble learning with label proportions for bankruptcy prediction

Z Chen, W Chen, Y Shi - Expert Systems with Applications, 2020 - Elsevier
Corporate bankruptcy prediction is an interesting and important research topic that can be
conceived in many practical applications. Recently, machine learning based methods have …

Object-based visual sentiment concept analysis and application

T Chen, FX Yu, J Chen, Y Cui, YY Chen… - Proceedings of the 22nd …, 2014 - dl.acm.org
This paper studies the problem of modeling object-based visual concepts such as" crazy
car" and" shy dog" with a goal to extract emotion related information from social multimedia …

Video event detection by inferring temporal instance labels

KT Lai, FX Yu, MS Chen… - Proceedings of the ieee …, 2014 - openaccess.thecvf.com
Video event detection allows intelligent indexing of video content based on events.
Traditional approaches extract features from video frames or shots, then quantize and pool …

On the minimal supervision for training any binary classifier from only unlabeled data

N Lu, G Niu, AK Menon, M Sugiyama - arXiv preprint arXiv:1808.10585, 2018 - arxiv.org
Empirical risk minimization (ERM), with proper loss function and regularization, is the
common practice of supervised classification. In this paper, we study training arbitrary (from …

On the domain adaptation and generalization of pretrained language models: A survey

X Guo, H Yu - arXiv preprint arXiv:2211.03154, 2022 - arxiv.org
Recent advances in NLP are brought by a range of large-scale pretrained language models
(PLMs). These PLMs have brought significant performance gains for a range of NLP tasks …

A survey on cost types, interaction schemes, and annotator performance models in selection algorithms for active learning in classification

M Herde, D Huseljic, B Sick, A Calma - IEEE Access, 2021 - ieeexplore.ieee.org
Pool-based active learning (AL) aims to optimize the annotation process (ie, labeling) as the
acquisition of annotations is often time-consuming and therefore expensive. For this …

Learning from label proportions: A mutual contamination framework

C Scott, J Zhang - Advances in neural information …, 2020 - proceedings.neurips.cc
Learning from label proportions (LLP) is a weakly supervised setting for classification in
which unlabeled training instances are grouped into bags, and each bag is annotated with …