Speaker: Thore Graepel When/Where: April 14, Wednesday, 15:00--16:00, Room 203 Title: Invariant Pattern Recognition by Semidefinite Programming Machines (Joint work with Ralf Herbrich) Abstract: Knowledge about local invariances with respect to given pattern transformations can greatly improve the accuracy of classification. Previous approaches are either based on regularisation or on the generation of virtual (transformed) examples. We developed a new framework for learning linear classifiers under known transformations based on semidefinite programming. We present a new learning algorithm -- the Semidefinite Programming Machine (SDPM) -- which is able to find a maximum margin hyperplane when the training examples are polynomial trajectories instead of single points. The solution is found to be sparse in dual variables and allows to identify those points on the trajectory with minimal real-valued output as virtual support vectors. Extensions to segments of trajectories, to more than one transformation parameter, and to learning with kernels are discussed. In experiments we use a Taylor expansion to locally approximate rotational invariance in pixel images from USPS and find improvements over known methods, which can be seen as approximations to the SDPM method. Bio: Thore Graepel received both his "Diplom" in physics and his PhD in Computer Science from the Technical University of Berlin. After postdoctoral positions at the Swiss Federal Institute of Technology Zurich and Royal Holloway College, London, he joined Microsoft Research Cambridge as a researcher in the Machine Learning and Perception group headed by Chris Bishop. His interests cover a wide range of problems in machine learning and inference. Thore is currently working on the application of machine learning to games ranging from computer games on Microsoft's Xbox to the ancient board game of Go.