Convolutional nets in practice
This lecture covers the concept of convolutional nets in practice and is a part of the Deep Learning Course at CDS, a course that covered the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition. Prerequisites for this course include: Introduction to Deep Learning and Introduction to Data Science or a Graduate Level Machine Learning.
00:00:00 – Welcome to class
00:00:09 – ConvNets in practice
00:01:49 – What are convolutions good for?
00:08:39 – Why do we need to stack layers?
00:13:50 – Object detection, multiple object recognition
00:17:25 – Multiple character recognition
00:19:31 – Sliding window ConvNet
00:23:20 – Face detection
00:25:55 – Whiteboard time!
00:31:30 – Q&A
00:33:40 – Semantic segmentation
00:38:54 – Robot navigation using semantic segmentation
00:43:42 – Category-level semantic segmentation
00:46:43 – FPGA ConvNet accelerator
00:47:56 – Error rate on ImageNet
00:49:23 – ResNet
00:50:36 – Networks comparison