TOCS Event-Silicon Valley Campus - Carnegie Mellon University

TOCS Event

Speaker:

George Vanecek
Sr. Principal Researcher/Architect, FutureWei Technologies, Center of Innovation

Date/Time:

February 12, 1:30 pm

Location:

Webcast:

CMUSV, Rm 118 [directions]

[login as guest]

Title: eCeNS Hybrid Knowledge Base and SCN Engine Framework for Building Intelligent Systems
Abstract:

For intelligent systems to sense, perceive and learn from their environment and the Internet, they need to create and manage the knowledge needed to perform their required tasks.  This talk introduces the eCeNS hybrid knowledge base and perception system being developed at the Innovation Center in Santa Clara, CA.  The system is being architected as a general-purpose framework for building domain specific intelligent systems with emphasis on ambient intelligence and organic computing.  The talk will also include a demonstration of the system's visual graphical editor and two POC use cases, one for home automation, and one for enterprise email-based context analysis. The talk will conclude with a more detailed presentation of the systems model for its world model, ontologies, concepts/taxonomies, and its neuron-inspired symbolic computational network. 

Speaker Bio:

George Vanecek, Jr., PhD, is a senior principal researcher/architect at FutureWei Technologies, Center of Innovation. His group is currently researching solutions for real-world perception, real-world situation modeling, context awareness and trust determination. George has extensive experience in designing and building software solutions, cloud applications and service platforms and has played key roles as a lead developer and chief architect in his ventures. Prior to FutureWei, he was a solutions architect at Cordys, a chief scientist at AT&T Internet Platforms Organization, and a research scientist at IBM and NIST. He founded several software start-ups after leaving academia as an Assistant Professor of Computer Science at Purdue University. He received his doctorate at University of Maryland in 1989 in areas of Artificial Intelligence and Geometric Modeling, where he designed and implemented a non-manifold solid modeling engine to perform near-real time contact analysis and collision detection to support the reasoning about physical objects for a process planning system and subsequently a rigid-body physical-based simulation system.