What is Deep Learning?
Machine Learning (ML) is a branch of Artificial Intelligence (AI) that focuses on teaching computers how to perform a given task by learning from data without explicitly being programmed to do so. Machine learning algorithms are trained on data to find patterns and correlations in the data and then use this analysis to make predictions and classifications. Applications of machine learning are today found in business, industry, government, and academia.
One of the most popular machine learning techniques is the Neural Network. Neural networks are loosely based on biological brains and consist of several layers of artificial neurons (simple CPUs). The neurons in one layer are connected by 'weights' to the neurons in the following layer. Weights are essentially real numbers and training the neural network to perform a useful task (prediction or classification) entails finding the right set of weights that will do the job.
Deep Learning is a revolutionary sub-branch of machine learning that started in 2006. Deep Learning can be described as the study, and application, of neural networks that have many (> 3) layers - hence why they are called 'deep'. Deep Learning has completely revolutionised pattern recognition, natural language processing, machine vision, and image processing. Modern deep learning algorithms are capable of performing tasks that were unthinkable only a decade ago. In the talk, we will discuss some of the basic ideas and concepts of AI and machine learning and then give an overview of deep learning, how it works, and what it is used for.