Predict 2019 has ended
avatar for Kevin Kuo

Kevin Kuo

Software Engineer
Talk: Uncertainty and Interpretability

Talk Description: In many business contexts, being able to quantify uncertainty of predictions and explain them are crucial to the success of predictive model implementations. We discuss industry case studies where these requirements come up and techniques and tools available to the data scientist to address them. Specifically, we demonstrate open source R packages that leverage probabilistic deep learning via TensorFlow and model explanation.

Biography: Kevin is a software engineer at RStudio building open source packages for machine learning, including the R interfaces to Spark and MLflow. Prior to RStudio, he held data science roles in various industries, including insurance, banking, and industrial manufacturing. Kevin is also the founder of Kasa AI, a community-driven initiative for open research in insurance analytics.

My Speakers Sessions

Tuesday, October 1

11:15am IST