The first part of the talk will explore issues with deep networks dealing with "unknowns" inputs, and the general problems of open-set recognition in deep networks. We review the core of open-set recognition theory and its application in our first attempt at open-set deep networks, "OpenMax" We discuss is successes and limitations and why classic "open-set" approaches don't really solve the problem of deep unknowns. We then present our recent work from NIPS2018, on a new model we call the ObjectoSphere. Using ObjectoSphere loss begins to address the learning of deep features that can handle unknown inputs. We present examples of its use first on simple datasets sets (MNIST/CFAR) and then onto unpublished work applying it to the real-world problem of open-set face recognition. We discuss of the relationship between open set recognition theory and adversarial image generation, showing how our deep-feature adversarial approach, called LOTS can attack the first OpenMax solution, as well as successfully attack even open-set face recognition systems. We end with a discussion of how open set theory can be applied to improve network robustness.
Dr. Terry Boult, El Pomar Professor of Innovation and Security at University of Colorado Colorado Springs (UCCS), does research in computer vision, machine learning, biometrics and security. Prior to joining UCCS in 2003, he was an endowed professor and founding chairman of Lehigh University's CSE Department and from 1986-1992 was a faculty at Columbia University. At University of Colorado at Colorado Springs he was the architecture of the awarding winning Bachelor of Innovation(tm) family of degrees and a key member in founding the UCCS Ph.d. in Security. Dr. Boult has been involved with multiple start up companies in the security space. He is innovator with passion for combining teaching, research and business. He has won multiple teaching, research, innovation and entrepreneurial awards and in 2017 was elected an IEEE Fellow. More details can be found at www.vast.uccs.edu/~tboult/vita.html