TensorFlow is a powerful open-source software library for machine learning developed by researchers at Google. It has many pre-built functions and models to ease the task of building and training neural networks, both on single machine and on multiple machines. TensorFlow also has dedicated modules to support the seamless integration of research with production.
This tutorial will cover the fundamentals and contemporary usage of the TensorFlow library for deep learning. It aims to help the audience understand the design choices that led to TensorFlow 1.0 and 2.0, explore the functions it has to offer, and learn how to build and structure models best suited for a deep learning project. Through the tutorials, students will use TensorFlow to build models of different
complexity, from simple linear/logistic regression to convolutional neural networks, recurrent neural networks, and self-attention models to solve tasks such as word embedding, translation, optical character recognition, joint intent slot filling for conversational AI. Students will also learn the best practices to structure a model and manage research experiments.
TopicsNatural Language Processing, Machine Learning, Artificial Intelligence
Vincenzo Sciacca is an experienced AI Technical Architect, responsible to enable and guide customers and Business partner in their high priority innovation initiatives, leveraging the Almawave proprietary Natural Language Processing technology known as IRIDE. He moved to Almawave after a longexperience with IBM covering different roles in SW development, Cloud and AI. He spent more than10 years in the Tivoli Laboratory an internationalIBM Research & Development focused on systems monitoring and management. He gained rich experienceand knowledge as Software Architect leading many International software project and as Data and AI Architect in the Cloud&Watson division. Vincenzo is also an active inventor, achieving multiple patents and authoring technical papers andhe always strived to apply AI/ML to traditional IT projects.