Admissible Machine Learning
AutoML serves as a means to explore various algorithms, mitigating the limitations of the no free lunch theorem, which suggests no single model is universally superior. Erin introduces the concept of admissible machine learning, a novel approach focusing on fairness and grounded in information theory, currently implemented exclusively in H2O. This emerging field promises to enhance model selection and performance across diverse scenarios.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
627: AutoML: Automated Machine Learning — with Erin LeDell
Related Questions