Bayesian Insights in ML
Rob emphasizes the complexity of priors in Bayesian statistics, arguing that what seems uninformative in one context may hold significance in another. Jon draws a fascinating parallel between initializing parameters in machine learning and the concept of priors, suggesting that the random values used in deep learning models serve a similar purpose. Both discuss the importance of soft constraints and the implications of parameter boundaries on model learning.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
SDS 507: Bayesian Statistics — with Rob Trangucci
Related Questions