What augmentations are sensitive to hyper-parameters and why?

Abstract

We apply augmentations to our dataset to enhance the quality of our predictions and make our final models more resilient to noisy data and domain drifts. Yet the question remains, how are these augmentations going to perform with different hyper-parameters? In this study we evaluate the sensitivity of augmentations with regards to the model's hyper parameters along with their consistency and influence by performing a Local Surrogate (LIME) interpretation on the impact of hyper-parameters when different augmentations are applied to a machine learning model. We have utilized Linear regression coefficients for weighing each augmentation. Our research has proved that there are some augmentations which are highly sensitive to hyper-parameters and others which are more resilient and reliable.

10 pages, 17 figures

References in corpus (8)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• On Hyperparameter Optimization of Machine Learning Algorithms: Theory and Practice
• The Effectiveness of Data Augmentation in Image Classification using Deep Learning
• Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
• Deep Image: Scaling up Image Recognition
• A Bayesian Data Augmentation Approach for Learning Deep Models
• Deep CNN Ensemble with Data Augmentation for Object Detection
• Anomaly Detection based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation