Title: Machine Learning Seminar(2016.10.18)
Author: CMAC
Date: 2016-10-10 (12:29)

Machine Learning Seminar

Date/Time: Oct. 18th, Tues., 2:00 ~ 4:00 PM

Location: Office of Prof. Choe(#405), Yonsei University

Speaker: Jun-Sang Eom

Affiliation: MathWorks Korea

Title: 빅데이터 처리를 위한 MATLAB 환경 소개

Topic: 국내외 MATLAB & SIMULINK 활용 사례 소개, MATLAB을 활용한 수학적 모델링, 전역최적화 기법, GPU, Cluster 초고속 컴퓨팅기법, Big data, large scale 문제 해결 기법


1. Mathematical Modeling with MATLAB

MATLAB enables you to build mathematical models for forecasting and optimizing the behavior of complex systems. In this session, we demonstrate how you can:

  • Develop models using data fitting and first-principles modeling techniques Simulate models and develop custom post processing routines
  • Generate reports that document models and simulation results

The session covers use of the MATLAB language, symbolic expressions, and prebuilt graphical tools for specific modeling tasks and other approaches you can use to develop models.

2. Global Optimization with MATLAB

Engineers, scientists, and financial analysts use optimization to find better solutions to their problems. This webinar will present MathWorks global optimization solutions for finding the best solution, or multiple good solutions, to problems that contain multiple maxima or minima, including problems that are nonsmooth or discontinuous. Product demonstrations will illustrate how you can use global search, genetic algorithm, simulated annealing, or direct search solvers to solve challenging real-world problems. Improving optimization execution speed using parallel computing will also be discussed.

3. Parallel Computing with MATLAB

Learn strategies and techniques for speeding up your MATLAB applications. Included are tips on how to optimize the performance of the MATLAB code itself and how to use the MATLAB family of products to take advantage of advances in hardware, such as multicore machines and computer clusters.

4. GPU Computing with MATLAB

Learn how MATLAB users can leverage NVIDIA GPUs to accelerate computationally intensive applications in areas such as image processing, signal processing, and computational finance. We show the GPU-enabled functionality in MATLAB and various add-on toolboxes, and demonstrate how you can integrate your own custom CUDA kernels into MATLAB. We also demonstrate how MATLAB supports CUDA kernel development by providing a high-level language and development environment for prototyping algorithms and incrementally developing and testing CUDA kernels.

5. Tackling Big data with MATLAB

Data is everywhere and every year we store more and more of it. Huge data-sets present an amazing opportunity for discovering new things about our world, about the products we make and about how people interact with them. However, big data-sets also present some real challenges. How do we understand them? How do we interrogate them? How do we even read them? This talk looks at the tools that MATLAB® provides for dealing with data-sets of all sizes.