Overview
Watch a keynote conference talk from NDC Porto 2024 where Jodie Burchell explores the critical question of trust in machine learning models, particularly large language models (LLMs). Delve into the complexities of assessing model performance and accuracy in an era dominated by "black box" algorithms. Learn about the various ways machine learning models can produce misleading results and discover methods to evaluate their truthfulness and reliability. Gain valuable insights into the challenges of validating model outputs, especially as LLMs become increasingly prevalent in everyday applications. Master techniques for navigating through the confusion surrounding model performance assessment and develop a better understanding of how to verify the accuracy of machine learning outputs.
Syllabus
Keynote: Can you trust your (large language) model? - Jodie Burchell - NDC Porto 2024
Taught by
NDC Conferences