Data Tech has ended
Use the event Sched to plan your day and rate the sessions you attend. View venue map
Back To Schedule
Thursday, May 30 • 2:15pm - 3:00pm
Statistical Optimization of Deep Learning Hyperparameters and Data Augmentation Methods

Log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
Training dataset sizes are a critical factor in the creation of accurate neural network image classifiers where the axiom “bigger-is-better” appears to hold true. Image augmentation (creating several instances of an original image by applying image transformations) is often used to increase the size of datasets when the original dataset is too small. Although several augmentation strategies have been developed that have successfully improved neural network performance, little research has been done to study the appropriate ratio of augmented to original data. In this presentation, Tom will introduce a search strategy that uses statistical mixture experiments to identify the optimal blend of several different image augmentation methods. He will also discuss how hyperparameter tuning can be incorporated into this process to simultaneously tune hyperparameters and augmentation strategies for efficient deep learning model optimization. The presentation concludes with a case study where a mixture experiment was used to identify the optimal augmentation strategy for a neural network used for manufacturing visual defect detection, resulting in a significant improvement in performance on a validation dataset.


Tom Albrecht

Principal Data Scientist, Boston Scientific
Tom Albrecht specializes in developing deep learning and natural language processing (NLP) models for high dimensional data, including image and free text data. He has also developed several linear and non-linear experimental design techniques. 

Thursday May 30, 2019 2:15pm - 3:00pm CDT
(D) P0808 A&B Normandale Partnership Center, 9700 France Ave So, Bloomington, MN 55431