bioRxivBioinformaticsPreprint
Gradient-based implementation of linear model outperforms deep learning models
Hamid HamidiiD, Dinghao WangiD, Quan LongiD, Matthew GreenbergiD
Abstract
Deep learning has been widely considered more effective than traditional statistical models in modeling biological complex data such as single-cell omics. Here we show the devil is hidden in details: by adapting a modern gradient solver to a traditional linear mixed model, we showed that conventional models can outperform deep models in terms of both speed and accuracy. This work reveals the potential of re-implementing traditional models with modern solvers.
- Citations
- 0
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Deep learning for brains?: Different linear and nonlinear scaling in UK Biobank brain images vs. machine-learning datasets
Marc-André Schulz, B.T. Thomas Yeo, … Danilo Bzdok·Nature CommunicationsPeer-reviewed·Aug 25, 2020
- A general deep hybrid model for bioreactor systems: combining first Principles equations with deep neural networks
José Pinto, Mykaella Mestre, … Rui Oliveira·Computers & Chemical EngineeringPeer-reviewed·Aug 9, 2022
- Modern machine learning outperforms GLMs at predicting spikes
Ari S. Benjamin, Hugo L. Fernandes, … Konrad Paul Kording·Frontiers in Computational NeurosciencePeer-reviewed·Jul 19, 2018
- Deep learning applications in single-cell omics data analysis
Nafiseh Erfanian, A. Ali Heydari, … Amirhossein Sahebkar·bioRxiv·Nov 26, 2021
- Deep learning in bioinformatics: introduction, application, and perspective in big data era
Yu Li, Chao Huang, … Xin Gao·MethodsPeer-reviewed·Apr 22, 2019
- Mini-batch optimization enables training of ODE models on large-scale datasets
Paul L. Stapor, Leonard Schmiester, … Jan Hasenauer·Nature CommunicationsPeer-reviewed·Jan 10, 2022