Artificial Intelligence

Creativity, Curiosity, Consistency, Controllability, and Complexity — The Cs of Cognitive Computing

Monday, December 11, 2017, 11:00am - 12:00pm PDTiCal
11th floor large conference room
This event is open to the public.
AI Seminar
Alex Schwing, University of Illinois at Urbana-Champaign

The remarkable advances of artificial intelligence algorithms have led to a widespread adoption in many areas, where they successfully support human judgement. Recent trends suggest that an increasing amount of available data, improvements regarding computational resources and more complex models will result in expert-level support by artificial intelligence in many more fields, particularly those where a single, accurate result is clearly available.

However, for cognitive computing tasks where we want to generate complex data and products, such as 3D models, videos, or missing regions in images, a single correct result does not exist. In contrast, many outcomes are viable. Algorithmic support in these areas generally relies on low-level operations with a confined goal, such as appearance based selection. As a consequence, a significant amount of domain expertise is required to operate those tools by linking a sequence of local low-level operations.

We develop algorithms which transform low-level human-computer interaction dominated workflows into a semi-automatic high-level manipulation using successively provided user preference feedback on automatically generated products. This form of operation will, for the first time, provide accessibility to image, video, and 3D model design for a large audience. To achieve this goal, we solve algorithmic challenges in the areas of `creativity,’ `curiosity,’ `consistency,’ `controllability,’ and `complexity.’  In this talk I’ll detail those five steps referred to as the five `Cs’ of cognitive computing, which are crucial components for a widely accessible workflow, tackling complex tasks such as editing of images, videos and 3D models.


Alex Schwing is an Assistant Professor at the University of Illinois at Urbana-Champaign working with talented students on computer vision and machine learning topics. He received is B.S. and diploma in Electrical Engineering and Information Technology from Technical University of Munich in 2006 and 2008 respectively, and obtained a PhD in Computer Science from ETH Zurich in 2014. Afterwards he joined University of Toronto as a postdoctoral fellow until 2016. His research interests are in the area of computer vision and machine learning, where he has co-authored numerous papers on topics in scene understanding, inference and learning algorithms, deep learning, image and language processing and generative modeling. His PhD thesis was awarded an ETH medal.



« Return to Events