06-17, 15:30–17:00 (Europe/London), Tower Suite 1
Exploring the process, implementation, practical applications, and advantages of stacking neural networks together. The tutorial focuses on building tunable, high-performance, multi-data-type feature models seamlessly using network concatenations in TensorFlow. We implement 3 examples and also derive explainability for a stacked neural network.
Stacking different Neural Nets together is extremely beneficial for applications that involve the availability of different data types ( Images + Text + Tabular ) to make better decisions. In this tutorial, we will find answers to fundamental questions behind stacking: Why stack? How to stack? Train - individually / together? Number of models? Hyperparameter tuning? Stacking pre-trained models? Accessing models post stacked training and very importantly, explainability? and consequently, implement Stacking for 3 use cases - Multi Data Type, Pretrained Model, Restricted Neural Nets.
Previous knowledge expected
Pranjal is an experienced AI Scientist building the first AI powered platform to accelerate R&D for Material Sciences across the globe. He loves opening black-box models to reveal insightful AI secrets that help decision makers adapt with the ever changing Industry needs. He also loves to teach and mentor passionate individuals aspiring to be a part of the Data Science Community, all with his favourite language, Python!