Within broad themes of Artificial Intelligence and Games, ASYNC is a series of low-maintenance pop-up events where sharing the most cutting edge, asynchronous, research leads to an international discussion and drives the field forward. The hope is that presenters at ASYNC events will get some interesting feedback on their formative work, novel collaborations will spring up and the community will gain much food for thought. And no one will have to work very hard!


ASYNC October theme:
Machine Learning and Games

The format will be a series of screenings of 15 minute video presentations, held during each afternoon (2pm-4pm UK time) of the week from 25th to 27th October, 2017. We will be announcing the schedule very soon.

We have set up a room for the screenings in TogetherTube here: MetaMakers Room.

There is no need to register, please just visit the screening room at the right time, watch the presentation and join in the chat! The presenters will be available for text questions during the screening and for a short while afterwards.

Schedule of video presentations

Wednesday 25th October

14:00 – 14:15
Niels Justesen, IT University of Copenhagen, Denmark.
Learning Macromanagement in StarCraft using Deep Learning (CIG’2017)

14:20 – 14:35
Michael Cook, MetaMakers Institute, Falmouth University, UK
The Chinese Room with a view: Why I haven’t used machine learning (yet)

14:40 – 14:55
Raluca Gaina, EECS, Queen Mary University of London, UK
Planning in Real-time Games with Rolling Horizon Evolution

15:00 – 15:15
Christopher Beckham, Montreal Institute of Learning Algorithms, Canada.
A step toward procedural terrain generation with generative adversarial networks

15:20 – 15:35
Adam Summerville, Expressive Intelligence Studio, UC Santa Cruz, USA.
Learning to read games — inductive logic programming for game analysis

15:40 – 16:00
Group discussion on chat.

Thursday 26th October

14:00 – 14:15
Daniel Karavolos, Institute of Digital Games, University of Malta.
When to use a shotgun: Linking weapons and maps through machine learning

14:20 – 14:35
Ed Powley, MetaMakers Institute, Falmouth University, UK
Emulating human play in a leading mobile card game

14:40 – 14:55
Phil Lopes, Computer Vision and Multimedia Lab, University of Geneva, Switzerland.
I know how you feel – Combining emotion recognition for level soundscape construction

15:00 – 15:15
Sam Snodgrass, Computer Science Department, Drexel University, USA.
Leveraging Multi-layer Representations for Procedural Level Generation

15:20 – 15:35
Kate Compton, UC Santa Cruz, USA.
Getting Lost in the Possibility Space: Adventures in Genetic Algorithms

15:40 – 16:00
Group discussion on chat.

Friday 27th October

14:00 – 14:15
Alexander Semenov, National Research University Higher School of Economics, Russia.
Machine learning for win prediction in Dota 2

14:20 – 14:35
Simon Colton, MetaMakers Institute, Falmouth University, UK
First thoughts on and first experiments in ML-based style transfer for game assets

14:40 – 14:55
Matthew Guzdial, Georgia Institute of Technology School of Interactive Computing, USA.
Toward ML-based Creative Automated Game Design

15:00 – 15:15
Ahmed Khalifa, NYU Game Innovation Lab, USA.
Using Evolutionary Algorithms to find new tree search algorithms

15:20 – 15:30
Group discussion on chat.