Introduction to Deep Learning - online

This is a hands-on introduction to the first steps in Deep Learning, intended for researchers who are familiar with (non-deep) Machine Learning.  This three day introduction aims to cover the basics of Deep Learning in a practical and hands-on manner, so that upon completion, you will be able to train your first neural network and understand what next steps to take to improve the model. 

The course covers:

• What is deep learning?

• Classification by a neural network using Keras

• Monitor the training progress

• Advanced layer types

• Real world application

By the end of the course participants will:

• Define deep learning
• Describe how a neural network is build up
• Explain the operations performed by a single neuron
• Describe what a loss function is
• Recall the sort of problems for which deep learning is a useful tool
• List some of the available tools for deep learning
• Recall the steps of a deep learning workflow
• Test that you have correctly installed the Keras, Seaborn and scikit-learn libraries
• Use the deep learning workflow to structure the notebook
• Explore the dataset using pandas and seaborn
• Identify the inputs and outputs of a deep neural network.
• Use one-hot encoding to prepare data for classification in Keras
• Describe a fully connected layer
• Implement a fully connected layer with Keras
• Use Keras to train a small fully connected network on prepared data
• Interpret the loss curve of the training process
• Use a confusion matrix to measure the trained networks’ performance on a test set
• Explain the importance of keeping your test set clean, by validating on the validation set instead of the test set
• Use the data splits to plot the training process
• Explain how optimization works
• Design a neural network for a regression task
• Measure the performance of your deep neural network
• Interpret the training plots to recognize overfitting
• Use normalization as preparation step for deep learning
• Implement basic strategies to prevent overfitting
• Understand why convolutional and pooling layers are useful for image data
• Implement a convolutional neural network on an image dataset
• Use a dropout layer to prevent overfitting
• Be able to tune the hyperparameters of a Keras model
• Adapt a state-of-the-art pre-trained network to your own dataset
• Understand that what we learned in this course can be applied to real-world problems
• Use best practices for organising a deep learning project
• Identify next steps to take after this course

IMPORTANT: Please note that this course includes computer workshops. Before registering please check that you will be able to access the software noted below. Please bear in mind minimum system requirements to run software and administration restrictions imposed by your institution or employer with may block the installation of software.

Pre-requisites:

Learners are expected to have the following knowledge:

• Basic Python programming skills and familiarity with the Pandas package.
• Basic knowledge on machine learning, including the following concepts: Data cleaning, train & test split, type of problems (regression, classification), overfitting & underfitting, metrics (accuracy, recall, etc.).