Session Chair:Ambarish Natu (General Co-Chair)
Room: Plaza Terrace
Deep learning has become the default method for solving most ML problems but has some significant limitations. These include the inherent difficulty in editing, updating, and controlling deep models. One of the causes of this problem is the fact that deep models store information implicitly in the values of multiple weights. This talk will cover two approaches to solving this problem by augmenting a deep model with explicit information storage. The resulting augmented models are able to be updated on the fly, at inference time, but also outperform standard deep learning architectures when this is not required. The approach also points towards methods that might enable late-binding in deep learning, which would broaden their applicability significantly.
Session Chair: Zhu Li, University of Missouri, Kansas City, USA.
Room: Plaza Terrace
Session Chair: Reji Matthew
Time: 12:00-13:30
Room: M9
Session Chair: Zhu Li, University of Missouri, Kansas City, USA.
Room: Plaza Terrace
View Presentation
Extended Realities aim to blend further our digital and analog worlds, these are immersive technologies that will be deeply connected to human perception and behaviour with their environment. Exploring them all together: the technology and the humans, will help to highlighting which things are particular to this medium. In this keynote we present a body of research done across the space of immersive technologies, from avatars to haptics, to authoring tools. This will be the starting point to discuss why is XR so interesting, and how it compares or interface to other mediums. All to find what is impossible to do outside of immersive technology, and how will AI impact this technology.
Session Chair: Reji Matthew
Time: 13:30 to 15:00 pm /15:30 to 17:00
Room: Booth Area
Read More
Time: 8:30 - 9:30 pm
Room: Booth Area
This talk presents a history of personal information devices. The origin is an audio player dated back to the 1990s which was born at an intersection of audio coding algorithms to provide sufficient subjective audio quality and a sufficient memory size on a single chip. LSI technology was indispensable to its birth which had a revolutionary impact on the hardware business. The audio-only device was naturally extended to include video signals to cover multimedia applications commonly encountered today in our daily life. Integration with a mobile phone brought us continuous extensions to wearables, connected operations, and sensing functions.
Session Chair: Zhu Li, University of Missouri, Kansas City, USA.
Room: Plaza Terrace
Time: 9:30am - 11:00am
Session Chair: Reji Matthew
Time: 13:30 to 15:00 pm /15:30 to 17:00
Room: Booth Area
Read More
Time:14:00-15:30pm
Room: Merivales Boardroom
