Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds

Prajit T. Rajendran; Huascar Espinoza; Agnes Delaborde; Chokri Mraidha

doi:10.1109/MLSP55844.2023.10285861

Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds

Prajit T. Rajendran^*
, Huascar Espinoza
, Agnes Delaborde
, Chokri Mraidha

^*Corresponding author for this work

List
KDT JU
Laboratoire national de métrologie et d'essais

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper proposes Stream-based Active Learning method using Adaptive uncertainty and Diversity Thresholds (SALAT), which aims to reduce the amount of data required for labeling in stream-based data settings while accounting for concept drift. The proposed method uses adaptive uncertainty and diversity thresholds, which are monitored by a sliding window of data selection history to ensure that the data selection is neither too conservative nor too aggressive. The effectiveness of SALAT is validated on the MNIST and FashionMNIST datasets, where it achieves similar performance to the baseline with fewer training samples. The experiments also confirm the hypothesis that a fixed threshold approach is deficient, and an adaptive threshold significantly reduces the data requirement to reach an adequate level of performance. The paper shows that SRLRT can handle concept drift effectively, making it applicable to various real-world applications such as text classification, image classification, and speech recognition.

Original language	English
Title of host publication	Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023
Editors	Danilo Comminiello, Michele Scarpiniti
Publisher	IEEE Computer Society
ISBN (Electronic)	9798350324112
DOIs	https://doi.org/10.1109/MLSP55844.2023.10285861
Publication status	Published - 2023
Externally published	Yes
Event	33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023 - Rome, Italy Duration: 17 Sept 2023 → 20 Sept 2023

Publication series

Name	IEEE International Workshop on Machine Learning for Signal Processing, MLSP
Volume	2023-September
ISSN (Print)	2161-0363
ISSN (Electronic)	2161-0371

Conference

Conference	33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023
Country/Territory	Italy
City	Rome
Period	17/09/23 → 20/09/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 4 Quality Education

Keywords

active learning
data selection
neural networks
stream-based data

Access to Document

10.1109/MLSP55844.2023.10285861

Cite this

Rajendran, P. T., Espinoza, H., Delaborde, A., & Mraidha, C. (2023). Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds. In D. Comminiello, & M. Scarpiniti (Eds.), Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023 (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2023-September). IEEE Computer Society. https://doi.org/10.1109/MLSP55844.2023.10285861

Rajendran, Prajit T. ; Espinoza, Huascar ; Delaborde, Agnes et al. / Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds. Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023. editor / Danilo Comminiello ; Michele Scarpiniti. IEEE Computer Society, 2023. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP).

@inproceedings{9424028f90c0473982143b5f3450d59d,

title = "Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds",

abstract = "This paper proposes Stream-based Active Learning method using Adaptive uncertainty and Diversity Thresholds (SALAT), which aims to reduce the amount of data required for labeling in stream-based data settings while accounting for concept drift. The proposed method uses adaptive uncertainty and diversity thresholds, which are monitored by a sliding window of data selection history to ensure that the data selection is neither too conservative nor too aggressive. The effectiveness of SALAT is validated on the MNIST and FashionMNIST datasets, where it achieves similar performance to the baseline with fewer training samples. The experiments also confirm the hypothesis that a fixed threshold approach is deficient, and an adaptive threshold significantly reduces the data requirement to reach an adequate level of performance. The paper shows that SRLRT can handle concept drift effectively, making it applicable to various real-world applications such as text classification, image classification, and speech recognition.",

keywords = "active learning, data selection, neural networks, stream-based data",

author = "Rajendran, \{Prajit T.\} and Huascar Espinoza and Agnes Delaborde and Chokri Mraidha",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023 ; Conference date: 17-09-2023 Through 20-09-2023",

year = "2023",

doi = "10.1109/MLSP55844.2023.10285861",

language = "English",

series = "IEEE International Workshop on Machine Learning for Signal Processing, MLSP",

publisher = "IEEE Computer Society",

editor = "Danilo Comminiello and Michele Scarpiniti",

booktitle = "Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023",

address = "United States",

}

Rajendran, PT, Espinoza, H, Delaborde, A & Mraidha, C 2023, Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds. in D Comminiello & M Scarpiniti (eds), Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023. IEEE International Workshop on Machine Learning for Signal Processing, MLSP, vol. 2023-September, IEEE Computer Society, 33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023, Rome, Italy, 17/09/23. https://doi.org/10.1109/MLSP55844.2023.10285861

Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds. / Rajendran, Prajit T.; Espinoza, Huascar; Delaborde, Agnes et al.
Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023. ed. / Danilo Comminiello; Michele Scarpiniti. IEEE Computer Society, 2023. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2023-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds

AU - Rajendran, Prajit T.

AU - Espinoza, Huascar

AU - Delaborde, Agnes

AU - Mraidha, Chokri

PY - 2023

Y1 - 2023

N2 - This paper proposes Stream-based Active Learning method using Adaptive uncertainty and Diversity Thresholds (SALAT), which aims to reduce the amount of data required for labeling in stream-based data settings while accounting for concept drift. The proposed method uses adaptive uncertainty and diversity thresholds, which are monitored by a sliding window of data selection history to ensure that the data selection is neither too conservative nor too aggressive. The effectiveness of SALAT is validated on the MNIST and FashionMNIST datasets, where it achieves similar performance to the baseline with fewer training samples. The experiments also confirm the hypothesis that a fixed threshold approach is deficient, and an adaptive threshold significantly reduces the data requirement to reach an adequate level of performance. The paper shows that SRLRT can handle concept drift effectively, making it applicable to various real-world applications such as text classification, image classification, and speech recognition.

AB - This paper proposes Stream-based Active Learning method using Adaptive uncertainty and Diversity Thresholds (SALAT), which aims to reduce the amount of data required for labeling in stream-based data settings while accounting for concept drift. The proposed method uses adaptive uncertainty and diversity thresholds, which are monitored by a sliding window of data selection history to ensure that the data selection is neither too conservative nor too aggressive. The effectiveness of SALAT is validated on the MNIST and FashionMNIST datasets, where it achieves similar performance to the baseline with fewer training samples. The experiments also confirm the hypothesis that a fixed threshold approach is deficient, and an adaptive threshold significantly reduces the data requirement to reach an adequate level of performance. The paper shows that SRLRT can handle concept drift effectively, making it applicable to various real-world applications such as text classification, image classification, and speech recognition.

KW - active learning

KW - data selection

KW - neural networks

KW - stream-based data

UR - https://www.scopus.com/pages/publications/85177162175

U2 - 10.1109/MLSP55844.2023.10285861

DO - 10.1109/MLSP55844.2023.10285861

M3 - Conference contribution

AN - SCOPUS:85177162175

T3 - IEEE International Workshop on Machine Learning for Signal Processing, MLSP

BT - Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023

A2 - Comminiello, Danilo

A2 - Scarpiniti, Michele

PB - IEEE Computer Society

T2 - 33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023

Y2 - 17 September 2023 through 20 September 2023

ER -

Rajendran PT, Espinoza H, Delaborde A, Mraidha C. Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds. In Comminiello D, Scarpiniti M, editors, Proceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023. IEEE Computer Society. 2023. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP). doi: 10.1109/MLSP55844.2023.10285861

Stream-Based Active Learning with Adaptive Uncertainty and Diversity Thresholds

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this