23. Convolutional Neural Networks

Vanilla neural networks are powerful, but convolutional neural networks are truly revolutionary! Instead of constructing features by hand, a convolutional neural network can extract features on its own! It does this through convolutional layers and then reduces dimensions for faster computing through pooling layers. This video describes how they work, dilated convolutions, kernel selection, stride, padding, max vs average pooling and more. Check out the whole materials informatics series at with workbooks and course notes available at 0:00 features in vanilla neural networks 2:10 position matters 5:05 what are convolutions and pooling layers? 8:55 feature extraction via kernels 15:20 simple to complex features via abstraction 17:05 different convolutions (dilation, padding) 21:40 pooling 23:00 fully connected layer to find patterns in features 25:27 visualization of CNNs in action!