{"id":459,"date":"2020-09-06T12:45:25","date_gmt":"2020-09-06T12:45:25","guid":{"rendered":"https:\/\/aaidm.buas.nl\/?page_id=459"},"modified":"2020-09-06T16:11:02","modified_gmt":"2020-09-06T16:11:02","slug":"ai1-preparation-for-week-6-more-on-back-propagation","status":"publish","type":"page","link":"https:\/\/aaidm.buas.nl\/?page_id=459","title":{"rendered":"AI1- Preparation for Week 6 More on Back-Propagation"},"content":{"rendered":"\n
\n
Author:<\/td>Francesco Lelli<\/a><\/td><\/tr>
Learning Line:<\/td>More on Back-propagation <\/td><\/tr>
Course:<\/td>AI1: Introduction to Neural Networks <\/td><\/tr>
Week:<\/td>6<\/td><\/tr>
Competencies:<\/td>1- Understand and be able to use neural network dynamics and factors that influence their performance
\u00ad2- Implement a basic pattern recognition<\/td><\/tr>
BoKS:<\/td>\u2022 3aS5 The student is able to develop a functional design for data driven applications.
\u2022 3aS2 The student knows how to apply the right statistical & mathematical tools for specific AI-contexts<\/td><\/tr>
Learning Goals:<\/td>Understanding Back-propagation and practical use cases of Neural Network<\/td><\/tr><\/tbody><\/table><\/figure>\n<\/div><\/div>\n\n\n\n

Artificial neural networks<\/strong> (ANN<\/strong>) or simply \u201cneural networks\u201d are computing systems inspired by \u201cclassical biological\u201d neural networks that are in animal brains. Typically an ANN \u201clearn\u201d to perform tasks by considering examples, without being programmed with task-specific rules. <\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Image recognition is a classical example where neural networks are used; in particular they learn to identify images that contain dogs by analyzing example images that have been manually categorized (labelled) as \u201cdog\u201d or \u201cno dog\u201d. Once the system \u201clearn\u201d the difference it can be used for identifying dogs in other images. <\/p>\n\n\n\n

An ANN does not use conceptual reasoning, for example is not capable to devise that \u201call dogs have 4 legs\u201d, or \u201ca dog must have a nose\u201d etc. Instead, they hide the features of a dog into their data structure and they \u201cinternally learn these characteristics\u201d from the examples that are used for training them.\u00a0<\/p>\n\n\n\n

Practical applications are more complex that the classical example mentioned in these videos. Nevertheless, learning how to recognize hand-written digits is always a good place to start. In other\u00a0words\u00a0you can consider this example as the \u201chello world\u201d example for learning the basics of Neural Networks.\u00a0<\/p>\n\n\n\n

In the past lecture we introduce back-propagation. During this week detail how back propagation works as well as we continue work to the assignment in order to produce our patter recognition. \u00a0<\/p>\n\n\n\n

\n