COMS 355 Chapter Notes - Chapter Glossary: Sebastopol, California, Ai Winter, Deep Learning
Document Summary
Deep learning is a subset of machine learning which is a field of artificial intelligence. It makes connections between different layers of data and processing them to make a computer able to create patterns and take decision in a given situation. To be effective, it has to be fed with an extremely large amount of data provided by humans. By putting data as inputs, we help the machine to adjust its algorithm. So, we give information to the machine to process, so it can be more effective in the creation of patterns, and more accurate in the decision-making process by making its algorithm as accurate as possible. The deep learning part refers to the process of hierarchization of these inputs (data). We are going to create different layers within the algorithm. Each layer is going to deal with a specific kind of data. This architecture is inspired by the structure of a brain.