Improved Single Target Accuracy Using Feature Selection for Multitarget Regression

Syed, Farrukh Hasan; Shah, Muhammad Raza; Tahir, Muhammad Atif

doi:10.1109/icetst49965.2020.9080703

Cited by 2 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method is not only a regression model (with a fixed graph) but also a model/feature selection strategy considering both accuracy and uncertainty, with the eventual task of determining an optimal graphical model on which to build our regressor. Such an approach has resemblance to classical methods in multi-target regressions (MTR) [45][46][47][48] , probabilistic graphical models (PGM) 49 , and uncertainty estimation learning 50,51 However, our problem setting is sufficiently different from usual statistical applications, and necessitates the development of novel techniques (Refer SI, section 3.5 for detailed discussion).…”

Section: Discussion -Beyond Validation and Broader Applications Of Th...mentioning

confidence: 99%

Machine Learning and High-Throughput Robust Design of P3HT-CNT Composite Thin Films for High Electrical Conductivity

Bash¹,

Cai²,

Chellappan³

et al. 2020

Preprint

View full text Add to dashboard Cite

<p>Combining high-throughput experiments with machine learning allows quick optimization of parameter spaces towards achieving target properties. In this study, we demonstrate that machine learning, combined with multi-labeled datasets, can additionally be used for scientific understanding and hypothesis testing. We introduce an automated flow system with high-throughput drop-casting for thin film preparation, followed by fast characterization of optical and electrical properties, with the capability to complete one cycle of learning of fully labeled ~160 samples in a single day. We combine regio-regular poly-3-hexylthiophene with various carbon nanotubes to achieve electrical conductivities as high as 1200 S/cm. Interestingly, a non-intuitive local optimum emerges when 10% of double-walled carbon nanotubes are added with long single wall carbon nanotubes, where the conductivity is seen to be as high as 700 S/cm, which we subsequently explain with high fidelity optical characterization. Employing dataset resampling strategies and graph-based regressions allows us to account for experimental cost and uncertainty estimation of correlated multi-outputs, and supports the proving of the hypothesis linking charge delocalization to electrical conductivity. We therefore present a robust machine-learning driven high-throughput experimental scheme that can be applied to optimize and understand properties of composites, or hybrid organic-inorganic materials.</p>

show abstract

Section: Discussion -Beyond Validation and Broader Applications Of Th...mentioning

confidence: 99%

Machine Learning and High-Throughput Robust Design of P3HT-CNT Composite Thin Films for High Electrical Conductivity

Bash¹,

Cai²,

Chellappan³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

Machine Learning and High-Throughput Robust Design of P3HT-CNT Composite Thin Films for High Electrical Conductivity

Bash¹,

Cai²,

Vijila³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

Improved Single Target Accuracy Using Feature Selection for Multitarget Regression

Cited by 2 publications

References 22 publications

Machine Learning and High-Throughput Robust Design of P3HT-CNT Composite Thin Films for High Electrical Conductivity

Machine Learning and High-Throughput Robust Design of P3HT-CNT Composite Thin Films for High Electrical Conductivity

Machine Learning and High-Throughput Robust Design of P3HT-CNT Composite Thin Films for High Electrical Conductivity

Contact Info

Product

Resources

About