Topological Data Analysis: potential applications to computer vision

Channel:
Microsoft Research
Subscribers:
348,000
Published on ● Video Link: https://www.youtube.com/watch?v=1EEVx4aVeU0


Category:
Vlog
Duration: 56:15
12,475 views
208

Topological Data Analysis quantifies hidden topological structures in big raw noisy data. The flagship tool (persistent homology) summarises the underlying structure across all scales. The stability result says that such a summary (persistence diagram) is robust to noise. We shall review potential approaches to combine topological data analysis and machine learning for practical problems in computer vision.



Other Videos By Microsoft Research


2016-06-22Towards Understandable Neural Networks for High Level AI Tasks - Part 5
2016-06-22IMS-Microsoft Research Workshop: Foundations of Data Science - False Discovery Rates - a new deal
2016-06-22Interactive Biotechnology: Cloud Labs, Biotic Games, DIY kits, and more
2016-06-22An Algorithm for Precision Medicine
2016-06-22Reverse Engineering Autonomous Language Acquisition
2016-06-22Code Hunt Workshop - Day 2 Session 7
2016-06-22NSF Interdisciplinary Workshop on Statistical NLP and Software Engineering - Session 4
2016-06-22Tutorial: High-Performance Hardware for Machine Learning
2016-06-22Representation Power of Neural Networks
2016-06-22Synthetic Biology: New Tools for an Industry at an Inflection Point
2016-06-22Topological Data Analysis: potential applications to computer vision
2016-06-22Tutorial: Large-Scale Distributed Systems for Training Neural Networks
2016-06-22Tutorial: Deep Learning
2016-06-22Towards Understandable Neural Networks for High Level AI Tasks - Part 4
2016-06-22Oral Session: Attractor Network Dynamics Enable Preplay and Rapid Path Planning
2016-06-22Python+Machine Learning tutorial - Introduction
2016-06-22DNA Strand Displacement
2016-06-22Design Expo 2015 - Pontifical Catholic University (Brazil)
2016-06-22Machine Learning Course - Lecture 16
2016-06-22Machine Learning Course - Lecture 7
2016-06-22Machine Learning Course - Lecture 17


Tags:
microsoft research
machine learning
deep neural networks