Sökning: onr:"swepub:oai:research.chalmers.se:6ac9f328-b8cf-404c-84e8-17a9c32dafce" >
Performance of Regr...
Performance of Regression Models as a Function of Experiment Noise
-
- Li, Gang, 1991 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Zrimec, Jan, 1981 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Ji, Boyang, 1983 (författare)
- Danmarks Tekniske Universitet,Technical University of Denmark,Chalmers tekniska högskola,Chalmers University of Technology
-
visa fler...
-
- Geng, Jun, 1985 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Larsbrink, Johan, 1982 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Zelezniak, Aleksej, 1984 (författare)
- Science for Life Laboratory (SciLifeLab),Chalmers tekniska högskola,Chalmers University of Technology
-
- Nielsen, Jens B, 1962 (författare)
- BioInnovation Institute (BII),Chalmers tekniska högskola,Chalmers University of Technology,Danmarks Tekniske Universitet,Technical University of Denmark
-
- Engqvist, Martin, 1983 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
visa färre...
-
(creator_code:org_t)
- 2021-06-27
- 2021
- Engelska.
-
Ingår i: Bioinformatics and Biology Insights. - : SAGE Publications. - 1177-9322. ; 15
- Relaterad länk:
-
https://research.cha... (primary) (free)
-
visa fler...
-
https://journals.sag...
-
https://research.cha...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- Background: A challenge in developing machine learning regression models is that it is difficult to know whether maximal performance has been reached on the test dataset, or whether further model improvement is possible. In biology, this problem is particularly pronounced as sample labels (response variables) are typically obtained through experiments and therefore have experiment noise associated with them. Such label noise puts a fundamental limit to the metrics of performance attainable by regression models on the test dataset. Results: We address this challenge by deriving an expected upper bound for the coefficient of determination (R2) for regression models when tested on the holdout dataset. This upper bound depends only on the noise associated with the response variable in a dataset as well as its variance. The upper bound estimate was validated via Monte Carlo simulations and then used as a tool to bootstrap performance of regression models trained on biological datasets, including protein sequence data, transcriptomic data, and genomic data. Conclusions: The new method for estimating upper bounds for model performance on test data should aid researchers in developing ML regression models that reach their maximum potential. Although we study biological datasets in this work, the new upper bound estimates will hold true for regression models from any research field or application area where response variables have associated noise.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Bioinformatik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Bioinformatics (hsv//eng)
- NATURVETENSKAP -- Matematik -- Sannolikhetsteori och statistik (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Probability Theory and Statistics (hsv//eng)
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik -- Reglerteknik (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering -- Control Engineering (hsv//eng)
Nyckelord
- machine learning
- label noise
- regression models
- upper bound
- experiment noise
Publikations- och innehållstyp
- art (ämneskategori)
- ref (ämneskategori)
Hitta via bibliotek
Till lärosätets databas
- Av författaren/redakt...
-
Li, Gang, 1991
-
Zrimec, Jan, 198 ...
-
Ji, Boyang, 1983
-
Geng, Jun, 1985
-
Larsbrink, Johan ...
-
Zelezniak, Aleks ...
-
visa fler...
-
Nielsen, Jens B, ...
-
Engqvist, Martin ...
-
visa färre...
- Om ämnet
-
- NATURVETENSKAP
-
NATURVETENSKAP
-
och Data och informa ...
-
och Bioinformatik
-
- NATURVETENSKAP
-
NATURVETENSKAP
-
och Matematik
-
och Sannolikhetsteor ...
-
- TEKNIK OCH TEKNOLOGIER
-
TEKNIK OCH TEKNO ...
-
och Elektroteknik oc ...
-
och Reglerteknik
- Artiklar i publikationen
-
Bioinformatics a ...
- Av lärosätet
-
Chalmers tekniska högskola