LMLow – Convolutional Neural Networks

Published on 9 February 2018

 

When: 3-4pm 23rd February 2018
Where: GIC, Roxby Building
How: please RSVP at the Event page

 

The fourth session of the “Learning Machine Learning Open Workgroup” (LMLow) will take place on February 23th. at 3pm in the GIC 6th. Floor, Roxby Building (turn left out of the lift). The session will cover Convolutional Neural Network (CNN) and image analysis, with an image classification example to demonstrate a real-world application of the method.

Any questions, comments or feedback, please get in touch with Dani at D.Arribas-Bel@liverpool.ac.uk and Mark at mark.green@liverpool.ac.uk.

 

Readings

 

For session 4 there are two primary readings and one optional reading. The first provides an overview from deep learning to CNN (to save your time just read pages 1-5 for the first), and the second elaborates what CNN is and how it works. For a deeper understanding, the third demonstrates an application of the method in image classification:

 

LeCun, Y., Bengio, Y. & Hinton, G. (2015). “Deep learning”. Nature 521, 436–444.

 

Standford University. (2017). Class CS231n: Convolutional Neural Networks (CNNs/ConvNets). [online] Available at: http://cs231n.github.io/convolutional-networks/ [Accessed 2 Feb. 2018].

 

Krizhevsky, A., Sutskever, I. & Hinton, G. (2012). “ImageNet classification with deep convolutional neural networks”. In Proc. Advances in Neural Information Processing Systems 25, 1090–1098.