Completed
Difficulty with defining objectives make AutoML challenging to apply in bioinformatics
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Multi-objective Bayesian Optimization with Heuristic Objectives for Biomedical and Molecular Data Analysis Workflows
Automatically move to the next video in the Classroom when playback concludes
- 1 Multi-objective Bayesian Optimization with Heuristic Objectives for Biomedical and Molecular Data Analysis Workflows
- 2 Overview
- 3 Bayesian optimization in bioinformatics has been applied to protein design
- 4 Most bioinformatic analyses are unsupervised
- 5 A typical workflow involves many steps and parameters
- 6 Difficulty with defining objectives make AutoML challenging to apply in bioinformatics
- 7 AutoML approaches construct objectives for a given problem
- 8 Motivation for AutoGeneS constructed objectives
- 9 Our method automatically infers which objectives are useful to guide optimization
- 10 MOBO basics
- 11 We build on the random scalarizations approach that returns a subset of the Pareto front
- 12 We determine the region of the Pareto front using objective behaviours
- 13 Three examples of desirable behaviours
- 14 Toy data simulating useful and not useful objectives
- 15 Optimizing cofactor in the analysis of Imaging Mass Cytometry (IMC) data
- 16 We construct objectives for clustering workflow using pairs of co-expressed proteins
- 17 We construct two meta-objectives using expert annotations
- 18 Parameters selected by our method led to clusterings that agree with expert annotations
- 19 Quantitative evaluation of performance
- 20 Summary
