Model validation using deep generation of stress data | Vitaliy Pozdnyakov | DSC Europe 23

In his tech tutorial, Vitaliy introduced a novel procedure for validating model robustness, emphasizing the use of deep generative models to sample unlikely events and introduce subtle shifts in input data. The approach provided a means to examine models' reactions to input data with varying likelihood levels. The tutorial underscored the increasing importance of model validation in the context of a growing number of running models, emphasizing the need to verify input and output data validity, assess model performance, ensure stability, and enhance interpretability. Attendees gained insights into leveraging deep generative models for robustness validation, contributing to a comprehensive understanding of model behavior under different conditions.

This tutorial by Vitaliy Pozdnyakov was held on November 21st as part of Tech Tutorials on Stream 5 live at the Data Science Conference Europe 2023.

Follow us on social media :
LinkedIn: Data Science Conference
https://www.linkedin.com/company/11184830
Instagram: @datasciconf
https://www.instagram.com/datasciconf/
Facebook page: @DataSciConference
https://www.facebook.com/DataSciConference/
Website: http://datasciconference.com/

Change the World Through Data

#datascience #ai #ml #artificialintelligence #bigdata #dataanalytics

Видео Model validation using deep generation of stress data | Vitaliy Pozdnyakov | DSC Europe 23 канала Data Science Conference