YoVDO

Deep Internal Learning - Deep Learning with Zero Examples

Offered By: Simons Institute via YouTube

Tags

Machine Learning Courses Data Science Courses Linguistics Courses Signal Processing Courses Acoustics Courses Image Segmentation Courses Super-Resolution Courses

Course Description

Overview

Explore the concept of Deep Internal Learning in this lecture from the Simons Institute's workshop on Decoding Communication in Non-Human Species. Delve into the innovative approach of learning from zero examples, as presented by Michal Irani from the Weizmann Institute. Discover how this technique can be applied to various fields, including super resolution, image segmentation, and even mind reading. Examine the differences between internal and external training methods, and analyze their respective results. Learn about the downscaling kernel, GAN applications, and the Deep Image Prior. Gain insights into how these advanced machine learning techniques can be utilized to decode communication in non-human animal and plant life, potentially revolutionizing our understanding and protection of diverse species on our planet.

Syllabus

Introduction
What is Deep Internal Learning
How can we learn
The two approaches
Super Resolution
ZeroShot CSSR
Internal vs External Training
Internal vs External Results
Downscaling Kernel
GAN
Results
Deep Image Prior
Image Segmentation
Mind Reading
Result
Questions


Taught by

Simons Institute

Related Courses

Sparse Representations in Image Processing: From Theory to Practice
Technion - Israel Institute of Technology via edX
Cutting Edge Deep Learning for Coders
Jeremy Howard via YouTube
Efficient Geometry-Aware 3D Generative Adversarial Networks - GAN Paper Explained
Aleksa Gordić - The AI Epiphany via YouTube
Beyond Text - Giving Stable Diffusion New Abilities
HuggingFace via YouTube
Single Image Super Resolution Using SRGAN
DigitalSreeni via YouTube