Lecture 5: Neural Networks

Lecture 5 introduces fully-connected neural networks as a powerful nonlinear classifier. We start by discussing feature transforms plus linear classifiers as mechanism for nonlinear classification, then introduce neural networks as a mechanism for jointly learning a feature transform and a classifier. We briefly discuss differences between biological and artificial neurons. We see how fully-connected neural networks perform nonlinear classification via space warping, and discuss the universal approximation property for neural networks. We end with a brief discussion of convexity: linear classifiers give rise to convex optimization problems which are more amenable to optimization than the nonconvex optimization problems required to learn neural network classifiers. Slides: http://myumi.ch/bvnX5 _________________________________________________________________________________________________ Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification and object detection. Recent developments in neural network approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of neural-network based deep learning methods for computer vision. During this course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. We will cover learning algorithms, neural network architectures, and practical engineering tricks for training and fine-tuning networks for visual recognition tasks. Course Website: http://myumi.ch/Bo9Ng Instructor: Justin Johnson http://myumi.ch/QA8Pg