2/4/2022: Review of: Auto-WEKA: Automatic Model Selection and Hyperparameter Optimization in WEKA, Serena Cheng, University of California, Berkeley

We present a paper that proposes a technique to solve the Combined Algorithm Selection and Hyperparameter Optimization (CASH problem). The idea is to automatically and simultaneously choose a learning algorithm and set its hyperparameters to optimize empirical performance. The authors formed the CASH problem as a single combined hierarchical hyperparameter optimization question, and through leveraging recent innovations in Bayesian Optimization, they proved this can be solved automatically. Specifically, they used Sequential Model-Based Optimization (SMBO) and Sequential Model-Based Algorithm Configuration (SMAC) using random forest to maximize positive expected improvement over existing given loss.

Source: Kotthoff, Lars, Chris Thornton, Holger H. Hoos, Frank Hutter, and Kevin Leyton-Brown. “Auto-WEKA: Automatic model selection and hyperparameter optimization in WEKA.” In Automated Machine Learning, pp. 81-95. Springer, Cham, 2019.

The presentation slides can be found here: Auto-WEKA presentation