Deep Learning with TensorFlow

Learn the fundamental concepts of neural networks and deep learning. One of the hottest fields in artificial intelligence, deep learning is the key to solving highly complex problems such as training autonomous vehicles or object recognition. New applications are being found every day across manufacturing, pharmaceutical, medical, security, transportation, and aerospace.

In this course, you will have the chance for hands-on exploration of TensorFlow, Google’s recently released in-house toolset for deep learning. It is a Python-based library that runs on graphics processing units and tensor processing units, executing trillions of instructions per second in parallel. A solid understanding of TensorFlow is critical for anyone working in fields involving AI and machine learning.

You will examine TensorFlow, GPU and TPU architectures, and machine learning approaches for applications in business intelligence and analytics, data mining, predictive maintenance, robotics behavior design, product development, marketing, operations improvement. When you complete the course, you will be prepared for advanced domains such as autonomy, internet of things, edge computing, cognitive computing/AI, computer vision, and natural language processing.

If you complete this course and either the Machine Learning for Advanced Analytics course or the Natural Language Processing course, you are eligible to receive the Caltech CTME Data Analytics Certificate.

Benefits

You will learn how to:

• Install TensorFlow software and access it via Python and R
• Understand the all Machine Learning (ML) Models
• Solve linear algebra related math problems in TensorFlow
• Understand Neural Network architecture and Deep Neural Networks
• Build Neural Networks models in TensorFlow
• Understand optimization algorithms - Gradient Descent + Adam – in TensorFlow
• Implement Backpropagation algorithm in TensorFlow
• Simulate Linear Regression, kNN, Clustering in TensorFlow
• Implement Convolution Neural Networks (CNN) in TensorFlow
• Implement Recurrent Neural Network (RNN) in TensorFlow
• Understand Reinforcement Learning
• Understand the role of TensorBoard in visualization

Topics

• Machine Learning + Neural Networks + Deep Learning
• Tools for Building Deep Learning
• TensorFlow Architecture
• Writing TensorFlow Programs
• Building Neural Networks in TensorFlow – Categorical and Numerical output
• Optimization - Gradient Descent, Adam, backpropagation
• Linear regression
• Convolution Neural Networks - Image classification
• Recurrent Neural Networks - Sequence to Sequence
• Reinforcement Learning

Who should attend

This program addresses the needs of developers, technologists, researchers, and engineers across a wide range of technology-driven industries. Participants should have a working knowledge of Python or R. You will examine TensorFlow and ML approaches for applications in business intelligence and analytics, data mining, predictive maintenance, robotics behavior design, product development, marketing, operations improvement. Participants are prepared for advanced domains such as autonomy, internet of things, edge computing, cognitive computing/AI, computer vision, and natural language processing.

Schedule

Classes are held in a Hybrid environment. This course now runs every Saturday from 8:00am to 12:00pm PT at the Caltech campus in Pasadena, CA. Since this class is now hybrid, participants also have the option of attending this class remotely via Zoom. There are 4 hours of instruction every Saturday for 6 Days, which makes a total of 24 instruction hours. During the week before each session you will receive emails containing location-specific instructions.

Fee: $2130 for the program. Payment is due (or credit card charged) on the Friday prior to first day of the course. You may pay the entire program fee of $2130 in one payment, or you may pay the program fee in 3 increments of $710.

Hours and Credits: Participants will earn 2.4 Continuing Education Units (CEUs).

Grading: These programs are based on a Pass/Fail grading system.

Instructor

Ash Pahwa, PhD