TY - JOUR
T1 - Stabilizing and Simplifying Sharpened Dimensionality Reduction using Deep Learning
AU - Espadoto, Mateus
AU - Kim, Youngjoo
AU - Trager, Scott
AU - Roerdink, Jos
AU - Telea, Alex
N1 - Funding Information:
This work is supported by the DSSC Doctoral Training Programme co-funded by the Marie Sklodowska-Curie COFUND project (DSSC 754315), and FAPESP under Grant 2020/13275-1, Brazil. The GALAH survey is based on observations made at the Australian Astronomical Observatory, under programmes A/2013B/13, A/2014A/25, A/2015A/19, A/2017A/18. We acknowledge the traditional owners of the land on which the AAT stands, the Gamilaraay people, and pay our respects to elders past and present. This work has made use of data from the European Space Agency (ESA) mission Gaia ( https://www.cosmos.esa.int/gaia ), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium ). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.
Publisher Copyright:
© 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.
PY - 2023/5
Y1 - 2023/5
N2 - Dimensionality reduction (DR) methods create 2D scatterplots of high-dimensional data for visual exploration. As such scatterplots are often used to reason about the cluster structure of the data, this requires DR methods with good cluster preservation abilities. Recently, Sharpened DR (SDR) was proposed to enhance the ability of existing DR methods to create scatterplots with good cluster structure. Following this, SDR-NNP was proposed to speed the computation of SDR by deep learning. However, both SDR and SDR-NNP require careful tuning of four parameters to control the final projection quality. In this work, we extend SDR-NNP to simplify its parameter settings. Our new method retains all the desirable properties of SDR and SDR-NNP. In addition, our method is stable vs setting all its parameters, making it practically a parameter-free method, and also increases the quality of the produced projections. We support our claims by extensive evaluations involving multiple datasets, parameter values, and quality metrics.
AB - Dimensionality reduction (DR) methods create 2D scatterplots of high-dimensional data for visual exploration. As such scatterplots are often used to reason about the cluster structure of the data, this requires DR methods with good cluster preservation abilities. Recently, Sharpened DR (SDR) was proposed to enhance the ability of existing DR methods to create scatterplots with good cluster structure. Following this, SDR-NNP was proposed to speed the computation of SDR by deep learning. However, both SDR and SDR-NNP require careful tuning of four parameters to control the final projection quality. In this work, we extend SDR-NNP to simplify its parameter settings. Our new method retains all the desirable properties of SDR and SDR-NNP. In addition, our method is stable vs setting all its parameters, making it practically a parameter-free method, and also increases the quality of the produced projections. We support our claims by extensive evaluations involving multiple datasets, parameter values, and quality metrics.
KW - High-dimensional visualization
KW - Dimensionality reduction
KW - Mean shift
KW - Neural networks
UR - http://www.scopus.com/inward/record.url?scp=85149227609&partnerID=8YFLogxK
U2 - 10.1007/s42979-022-01661-5
DO - 10.1007/s42979-022-01661-5
M3 - Article
SN - 2662-995X
VL - 4
JO - SN Computer Science
JF - SN Computer Science
IS - 3
M1 - 244
ER -