3330 - Advanced Model Optimization

Issued by GAIn® - The Global AI network

When building a machine learning model, you strive for optimal performance, but achieving this can be a cumbersome task. In this training, you get hands-on experience with advanced techniques on hyperparameter tuning, feature selection, and model validation. To mention a few, we will cover Bayesian optimization, genetic algorithms, elastic nets, and the proper way to apply cross-validation. Moreover, you will learn how to deal with data shifts and make your model more robust and future-proof.

Skills

Earning Criteria

Hyperparameter tuning – Master hyperparameter tuning to deliver maximum performance.
Feature selection – Automate the feature selection procedure for a faster modeling process and better performance.
Robustness - Evaluating and increasing a model’s robustness for stable performance with minimal overfit, for both common and rare cases.
Data shifts - Being able to identify and cope with data shifts, to ensure a future-proof model in production.
Exam