Talk: Automatic Machine Learning (AutoML)

About this event

Level: Beginner
Topic: Machine Learning, Auto ML, Neural Architecture Search(NAS)
Speaker: Hieu Dang

Machine learning research is recently exploding and advancing in many application domains, such as robotics, computer vision and speech, finance, cybersecurity, online commercials, etc. However, the performance of machine learning techniques is too sensitive to the experts’ design decisions of model architecture and hyperparameters. This is a noticeable barrier to new users to adopt machine learning to solve their problems.

Automatic machine learning (AutoML) is an emerging field of machine learning that targets to help users adopt machine learning with the optimal design decision and parameter configuration in an automatic way. With AutoML, the users just need to provide input data and problem specification, AutoML determines the optimal approach, model architecture, and hyperparameters for the best performance.

This talk addresses the changing world of automatic machine learning, from state of research to practical applications. The talk will cover three major issues. First, the talk starts from a review of the state of the art in automatic machine learning research, specifically in HyperParameter Optimization(HPO), Neural architect search (NAS), and Meta Learning. Second, it reviews open-source developments for AutoML. Finally, it provides practical examples of using AutoML.

Schedule:
6:30-7:00: Networking, snacks, and refreshments
7:00-7:45: Automatic Machine Learning (AutoML)
7:45-8:00: Q&A
8:00-8:30: Networking

About our speaker:

Hieu Dang currently works as a Senior Machine Learning Engineer at SkipTheDishes Inc. Prior to working at Skipthedishes, he was a Senior Machine Learning Engineer at Sightline Innovation Inc. He has worked on many interesting industrial data driven projects applying machine learning algorithms and signal processing techniques to derive insights from big data which require a lot of data wrangling, feature engineering, feature learning, and optimizing machine learning models. He obtained his PhD in electrical and computer engineering from the University of Manitoba, Winnipeg in 2015. His research interests include deep learning, probabilistic graphical models, nonlinear optimization, stochastic processes, and signal and image processing. He has authored and co-authored 16 refereed publications (patents, journal papers, conference papers) in his research areas.

The venue will be at the brand new Social Hub Space at SkipTheDishes, located on the main floor of 140 Bannatyne Avenue, Winnipeg.

We hope to see you there!
GDG Winnipeg


Organizers