Hyperparameter Optimization of Topological Features for Machine Learning Applications
- Resource Type
- Conference
- Authors
- Motta, Francis; Tralie, Christopher; Bedini, Rossella; Bini, Fabiano; Bini, Gilberto; Eramian, Hamed; Gameiro, Marcio; Haase, Steve; Haddox, Hugh; Harer, John; Leiby, Nick; Marinozzi, Franco; Novotney, Scott; Rocklin, Gabe; Singer, Jed; Strickland, Devin; Vaughn, Matt
- Source
- 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA) Machine Learning And Applications (ICMLA), 2019 18th IEEE International Conference On. :1107-1114 Dec, 2019
- Subject
- Computing and Processing
Engineering Profession
Robotics and Control Systems
Signal Processing and Analysis
Three-dimensional displays
Optimization
Predictive models
Data models
Machine learning
Bayes methods
Transforms
hyperparameter optimization
topological data analysis
persistence diagrams
machine learning
- Language
This paper describes a general pipeline for generating optimal vector representations of topological features of data for use with machine learning algorithms. This pipeline can be viewed as a costly black-box function defined over a complex configuration space, each point of which specifies both how features are generated and how predictive models are trained on those features. We propose using state-of-the-art Bayesian optimization algorithms to inform the choice of topological vectorization hyperparameters while simultaneously choosing learning model parameters. We demonstrate the need for and effectiveness of this pipeline using two difficult biological learning problems, and illustrate the nontrivial interactions between topological feature generation and learning model hyperparameters.