Neural Networks Using Tensorflow
Get an introduction to the core concepts needed to implement a neural network application in Tensorflow. A final project crystallizes key concepts and familiarizes students with Tensorflow protocols.
About this course:Neural networks have gained widespread recognition for their ability to provide solutions to applications for which alternative machine learning approaches are inadequate. A plethora of successful high profile applications has heightened public interest in this field. Successful applications include automated control of self-driving vehicles, computer chess and go algorithms that defeat all human opponents, automated language translation services, speech recognition, speech generation, as well as computer-generated music composition and artistic renderings. In the past, the field of neural networks was accessible to only specialists in the field. However, availability of software including Google’s recent release of its neural network software package, Tensorflow, allows practitioners in data science to configure, train, and deploy neural networks for a wide range of applications. This course introduces practitioners to the core concepts needed to make key architectural and configuration decisions and then implement a neural network application in Tensorflow. A final project crystallizes key concepts as well as familiarizing the student with Tensorflow protocols.
Programming: The student is expected to have basic programming skills that one would obtain from experience with a high-level language. Knowledge of Python would be helpful but is not required. Students without Python experience are welcome and must be prepared to learn basic syntax and control structures on their own.
Mathematics: As Tensorflow performs the more complicated mathematical operations, mathematical prerequisites are minimal; precalculus is the only prerequisite. Students with a stronger background will have the opportunity to investigate topics in more depth and will find some material more accessible.
Data Science: Students should be familiar with basic tools of data science including preparing data, sampling, and regression.