Problem Formulation for Machine Learning with Romer Rosales - TWiML Talk #149

This article summarizes a podcast episode featuring Romer Rosales, Director of AI at LinkedIn. The discussion covers graphical models, approximate probability inference, and the application of machine learning at LinkedIn. A key focus is on problem formulation and selecting appropriate objective functions, highlighting LinkedIn's 'holistic approach' to ML projects. The conversation also touches upon tools developed to scale data science efforts, such as optimization solvers and hyperparameter optimization. The episode promises an engaging discussion on practical aspects of machine learning.

• •The episode focuses on problem formulation and objective function selection in machine learning.
• •It highlights LinkedIn's 'holistic approach' to ML projects.
• •The discussion includes tools for scaling data science efforts, such as optimization solvers.