Due to the rapid spread of coronavirus (COVID-19) and resulting travel restrictions, Georgia Tech students and faculty will now be presenting their research virtually at the International Conference on Learning Representations (ICLR), one of the biggest artificial intelligence (AI) conferences in the world, April 25 through 30.
With 17 papers to present, researchers will create a five-minute video to be used during virtual poster sessions.
Research includes a paper co-written by Georgia Tech and Facebook about a new method for speeding up AI training by culling weak AI agents. Together, they have created an AI system that can navigate a 3D environment from a starting point to a goal with a 99.9% success rate and few mistakes.
“While I’m interested to see how the conference transitions to a virtual format, I am happy that the carbon footprint of people travelling and the cost of attendance will be greatly reduced. This may serve as a useful test case for organizers to support the rapid growth in attendance seen by multiple annual machine learning conferences,” said Jiachen Yang, ML@GT Ph.D. student.
Yang will be presenting his work on teaching multiple AI agents to work together to accomplish goals more quickly and accurately.
Devi Parikh, an associate professor in the School of Interactive Computing and the Machine Learning Center at Georgia Tech (ML@GT), will deliver the keynote address. Parikh will talk about AI systems at the intersection of computer vision and natural language processing. She will also give an overview of why problems at the intersection of vision and language are popular, what capabilities today’s AI systems have, and what challenges remain.
ICLR was scheduled to be the first premier artificial intelligence research conference to take place in Africa, being held in Addis Ababa, Ethiopia April 25-30.
For more information on Georgia Tech’s accepted work, visit http://bit.ly/ICLR2020.