[Recruitment]
Associate/Assistant Professor in Artificial Intelligence for Space-Enabled Technologies, Durham University

We are looking for applicants in Artificial Intelligence, Computer Vision, Edge Computing, Digital Twins, Human Computer Interaction, User Modelling, Robotics or Resilient Computing with potentials/achievements in informing space applications.

The post hoder will enjoy 1) a permanent (equivalent to US tenured) position at a top 100 university, 2) significantly reduced teaching, 3) a fully-funded PhD, 4) travel budget, 5) chance for a 2-year fully-funded Post-Doc.

Manifold Regularized Experimental Design for Active Learning

Lining Zhang, Hubert P. H. Shum and Ling Shao
IEEE Transactions on Image Processing (TIP), 2017

REF 2021 Submitted Output Impact Factor: 10.6 Top 10% Journal in Computer Science, Artificial Intelligence

Manifold Regularized Experimental Design for Active Learning

Abstract

Various machine learning and data mining tasks in classification require abundant data samples to be labeled for training. Conventional active learning methods aim at labeling the most informative samples for alleviating the labor of the user. Many previous studies in active learning select one sample after another in a greedy manner. However, this is not very effective because the classification models have to be retrained for each newly labeled sample. Moreover, many popular active learning approaches utilize the most uncertain samples by leveraging the classification hyperplane of the classifier, which is not appropriate since the classification hyperplane is inaccurate when the training data are small-sized. The problem of insufficient training data in real-world systems limits the potential applications of these approaches. This paper presents a novel method of active learning called manifold regularized experimental design (MRED), which can label multiple informative samples at one time for training. In addition, MRED gives an explicit geometric explanation for the selected samples to be labeled by the user. Different from existing active learning methods, our method avoids the intrinsic problems caused by insufficiently labeled samples in real-world applications. Various experiments on synthetic datasets, the Yale face database and the Corel image database have been carried out to show how MRED outperforms existing methods.

Downloads

YouTube

Citations

BibTeX

@article{zhang17manifold,
 author={Zhang, Lining and Shum, Hubert P. H. and Shao, Ling},
 journal={IEEE Transactions on Image Processing},
 title={Manifold Regularized Experimental Design for Active Learning},
 year={2017},
 month={2},
 volume={26},
 number={2},
 pages={969--981},
 numpages={14},
 doi={10.1109/TIP.2016.2635440},
 issn={1057-7149},
 publisher={IEEE},
}

RIS

TY  - JOUR
AU  - Zhang, Lining
AU  - Shum, Hubert P. H.
AU  - Shao, Ling
T2  - IEEE Transactions on Image Processing
TI  - Manifold Regularized Experimental Design for Active Learning
PY  - 2017
Y1  - 2 2017
VL  - 26
IS  - 2
SP  - 969
EP  - 981
DO  - 10.1109/TIP.2016.2635440
SN  - 1057-7149
PB  - IEEE
ER  - 

Plain Text

Lining Zhang, Hubert P. H. Shum and Ling Shao, "Manifold Regularized Experimental Design for Active Learning," IEEE Transactions on Image Processing, vol. 26, no. 2, pp. 969-981, IEEE, Feb 2017.

Supporting Grants

The Engineering and Physical Sciences Research Council
Interaction-based Human Motion Analysis
EPSRC First Grant Scheme (Ref: EP/M002632/1): £123,819, Principal Investigator ()
Received from The Engineering and Physical Sciences Research Council, UK, 2014-2016
Project Page

Similar Research

Lining Zhang, Hubert P. H. Shum and Ling Shao, "Discriminative Semantic Subspace Analysis for Relevance Feedback", IEEE Transactions on Image Processing (TIP), 2016

 

 

Last updated on 24 February 2024
RSS Feed