{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"omics_type":["Unknown"],"submitter":["Zhou J"],"funding":["NHGRI NIH HHS","NIH HHS","NIGMS NIH HHS"],"pubmed_abstract":["Deep neural networks (DNNs) have advanced predictive modeling for regulatory genomics, but challenges remain in ensuring the reliability of their predictions and understanding the key factors behind their decision making. Here we introduce DEGU (Distilling Ensembles for Genomic Uncertainty-aware models), a method that integrates ensemble learning and knowledge distillation to improve the robustness and explainability of DNN predictions. DEGU distills the predictions of an ensemble of DNNs into a single model, capturing both the average of the ensemble's predictions and the variability across them, with the latter representing epistemic (or model-based) uncertainty. DEGU also includes an optional auxiliary task to estimate aleatoric, or data-based, uncertainty by modeling variability across"],"journal":["bioRxiv : the preprint server for biology"],"pagination":["2024.11.13.623485"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC11601481"],"repository":["biostudies-literature"],"pubmed_title":["Uncertainty-aware genomic deep learning with knowledge distillation."],"pmcid":["PMC11601481"],"funding_grant_id":["R01 GM149921","S10 OD028632","R01 HG012131"],"pubmed_authors":["Tang Z","Koo PK","Zhou J","Rizzo K"],"additional_accession":[]},"is_claimable":false,"name":"Uncertainty-aware genomic deep learning with knowledge distillation.","description":"Deep neural networks (DNNs) have advanced predictive modeling for regulatory genomics, but challenges remain in ensuring the reliability of their predictions and understanding the key factors behind their decision making. Here we introduce DEGU (Distilling Ensembles for Genomic Uncertainty-aware models), a method that integrates ensemble learning and knowledge distillation to improve the robustness and explainability of DNN predictions. DEGU distills the predictions of an ensemble of DNNs into a single model, capturing both the average of the ensemble's predictions and the variability across them, with the latter representing epistemic (or model-based) uncertainty. DEGU also includes an optional auxiliary task to estimate aleatoric, or data-based, uncertainty by modeling variability across","dates":{"release":"2024-01-01T00:00:00Z","publication":"2024 Nov","modification":"2026-06-02T03:21:27.995Z","creation":"2025-04-03T23:56:11.721Z"},"accession":"S-EPMC11601481","cross_references":{"pubmed":["39605624"],"doi":["10.1101/2024.11.13.623485"]}}