AutoML Micro Analysis

In this study, we focus on the Microlevel by empirically evaluating and analyzing the performance impact of several design decisions and parameters including meta-learning,ensembling,time budget and size of search space,separately and combined. The results of our study reveal various interesting insights that can significantly guide and impact the design of AutoML frameworks.

• Benchmark Datasets
• Experiment Setup
• Results
  • Impact of Meta-learning
  • Impact of Ensembling
  • Impact of Combined Meta-Learning and Ensembling
  • Impact of Time Budget
  • Impact of Search Space
• Download Logs