Deep Neural Networks (DNNs) are used as a machine learning method for both regression and classification problems. TensorFlow is a popular software library that is often used to construct and train DNNs. In this workshop, participants will learn how to quickly use the high-level TensorFlow Estimator module to perform nonlinear regression and classification with standard fully connected DNNs. We will also show how the Estimator module can be used to perform image classification using Convolutional Neural Networks (CNNs). All examples will use Python; some familiarity with Python is recommended. You are encouraged but not required to bring a laptop with TensorFlow already installed.
Neural networks are powerful tools that output predictions given data; interesting areas of application include image recognition and autonomous driving.
In this workshop we introduce the basic concept of neural networks and demonstrate the use of neural networks using “PyTorch.”
PyTorch is one of the efficient Python packages for neural networks, which is designed to be highly flexible and intuitive.
A basic programming experience in Python is helpful to follow the PyTorch examples. Attendees are also welcome to follow the PyTorch examples on their own laptops during the workshop, in which case it is recommended that PyTorch be installed in advance; see https://github.com/pytorch/
This workshop is not intended to be comprehensive in terms of kinds of neural networks, optimization algorithms, or neural network frameworks, but rather it is intended to present an overview of conceptual and practical aspects of neural networks.
