题目: 1. Architectures for the FPGA Implementation of Online Kernel Methods
2. Measuring Sensitivity to Rounding Error using Monte Carlo Arithmetic
报告人：Philip Leong, The University of Sydney
时间：2017年4月17日（周一）下午1:303:30
地点：张江校区微电子楼269室
Abstract
1. In machine learning, traditional linear prediction techniques are well understood and methods for their efficient solution have been developed. Many realworld applications are better modelled using nonlinear techniques, which often have high computational requirements. Kernel methods utilise linear methods in a nonlinear feature space and combine the advantages of both. They are considered one of the major recent advances in machine learning research. Commonly used kernel methods include the support vector machine (SVM), Gaussian processes and regularisation networks. These are batchbased, and a global optimisation is conducted over all input exemplars to create a model. In contrast, online methods, such as the kernel recursive least squares (KRLS) algorithm, update the state in a recursive and incremental fashion upon receiving a new exemplar. Although not as extensively studied as batch methods, online approaches are advantageous when throughput and latency are critical.
In this talk I will describe efforts in the Computer Engineering Laboratory to produce highperformance implementations of online kernel methods. These have included: (1) a microcoded vector processor optimised for kernel methods; (2) a fully pipelined implementation of kernel normalised least mean squares which achieves 160 GFLOPS; (3) an implementation of Naive Online regularised Risk Minimization Algorithm (NORMA) which uses "braiding" to resolve data hazards and reduce latency by an order of magnitude; and (4) utilising random projections to make a low rank approximation to the input before processing.
2. Runtime analysis provides an effective method for measuring the sensitivity of programs to rounding errors. To date, implementations have required significant changes to source code, detracting from their widespread application. In this work we present an open source system that automates the quantitative analysis of floating point rounding errors, through the use of Cbased sourcetosource compilation and a Monte Carlo Arithmetic library. We demonstrate its application to the comparison of algorithms, detection of catastrophic cancellation, and determination of whether singleprecision floating point provides sufficient accuracy for a given application. Methods for obtaining quantifiable measurements of sensitivity to rounding error are also detailed.
Biography
Philip Leong received the B.Sc., B.E. and Ph.D. degrees from the University of Sydney. In 1993 he was a consultant to ST Microelectronics in Milan, Italy working on advanced flash memorybased integrated circuit design. From 19972009 he was with the Chinese University of Hong Kong. He is currently Acting Head of School and Professor of Computer Systems in the School of Electrical and Information Engineering at the University of Sydney, Visiting Professor at Imperial College, Visiting Professor at Harbin Institute of Technology, and Chief Technology Advisor to Cluster Technology.
