Donation

Sydney Machine Learning: Generative Models Special Edition, Wed, 30th of M...

Event Information

Share this event

Date and Time

Location

Location

Amazon, Level 37, 2 Park Street Sydney

Australia

View Map

Refund Policy

Refund Policy

Refunds up to 1 day before event

Event description

Description

We are lucky enough to have Amazon as our host for this Sydney Machine Learning meetup. Come along for beer, pizza, networking and talks from Dr Alan Blair, Head of the Machine Learning group at UNSW & Dr Sachin Abeywardana author of Deepschool.io for two extremely interesting talks on Generative Models.

Generative Adversarial Networks (GANs) is one method for performing unsupervised tasks with Neural networks & have had massive exposure since their invention by Ian Goodfellow in 2014 with applications being found in such crazy areas as drug discovery & text translation into direct images!!

GANs however are only one model type of a much broader field known as Generative Models. SML is honoured to have Dr Blair, who is going to give you a broad overview of his work in the field, & Dr Abeywardana, who will provide a live demo of the technology.


Dr Alan Blair’s Talk:

In this talk I will briefly summarize both historical and recent developments in adversarial and coevolutionary machine learning, including virtual creatures, sorting networks, game playing and computational creativity. I will describe some recent work in artist-critic coevolution, using a genetic program artist trained by hierarchical evolution and a deep convolutional neural network critic trained by gradient descent. Driven by selective pressure for low algorithmic complexity, combined with the need to fool the critic, the artist evolves to produce geometric shapes and patterns which remind us of everyday objects, landscapes and designs in a manner comparable to abstract art of the early 20th century. The talk will conclude with a discussion of recent developments in generative adversarial imitation learning and text processing, and possible future directions.

Alan Blair completed his PhD at MIT and worked at Brandeis, UQ and UniMelb, before taking up his current position at UNSW. His research interests include self learning for strategic games, robot navigation, language processing, convolutional network architectures and training, adversarial and coevolutionary dynamics, multi-task learning, hierarchical evolution and computational creativity.

Dr Blair’s personal website can be found at www.cse.unsw.edu.au/~blair


Dr Sachin Abeywardana’s Talk:

GANs are a completely new area within Deep Learning that aims to generate data. It is similar to Bayesian methods in the sense that it seeks to find the distribution of the data. In this talk I will talk about some of the basic theory and go through a python implementation of GANs. This is also one of the lessons offered in DeepSchool.io

Sachin graduated from the University of Sydney with an Undergraduate degree in Mechatronics Engineering and continued on to do his PhD in Bayesian Machine Learning. He has since worked as a Data Scientist in a Cyber Security company. He is the author of DeepSchool.io, an open source and free framework for teaching Deep Learning to software engineers and currently has 1000+ stars on GitHub.

Dr Abeywardana’s Deepschool.io can be found here on github


Schedule:

6:00 - 6:30: Beer, Pizza, Networking

6:30 - 6:40: Intro

6:40 - 7:10: Dr Blair's Talk

7:10 - 7:20: Small break

7:20 - 7:50 Dr. Abeywardana’s Talk

7:50 - 8:00 Wrap up

_______________________________________

Date and Time

Location

Amazon, Level 37, 2 Park Street Sydney

Australia

View Map

Refund Policy

Refunds up to 1 day before event

Save This Event

Event Saved