Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Dive into the fundamentals of neural networks in this comprehensive lecture from the University of Central Florida's CAP5415 course. Explore the challenges of object classification, pixel-based representation, and the importance of feature representations in image recognition. Examine the spectrum of supervision in machine learning and gain insights into the biological inspiration behind artificial neural networks. Learn about the computational implementation of neural activation functions, binary classification of images, and multi-class neural networks. Discover the concept of bias convenience and composition in neural network architectures. Address the limitations of linear functions in neural networks and understand why non-linear activation functions are crucial for complex problem-solving.
Syllabus
Intro
Our goal in object classification
Pixel-based representation
What we want
Some feature representations
Image classification - ImageNet
Features
Recognition task and supervision
Spectrum of supervision
The machine learning framework
Neurons in the Brain
Background in Neural Nets (NN)
Brain is a remarkable Computer
Computational Implementation of the Neural Activation Function
Binary classifying an image
Neural Networks - multiclass
Bias convenience
Composition
Problem 1 with all linear functions
Taught by
UCF CRCV