|
課程名稱 |
統計計算
Statistical computing |
|
開課學期 |
104-1 |
|
授課對象 |
理學院 數學系 |
|
授課教師 |
陳定立 |
|
課號 |
MATH5014 |
|
課程識別碼 |
221 U6710 |
|
班次 |
|
|
學分 |
3 |
|
全/半年 |
半年 |
|
必/選修 |
選修 |
|
上課時間 |
星期二6(13:20~14:10)星期四3,4(10:20~12:10) |
|
上課地點 |
天數101新103 |
|
備註 |
總人數上限：80人 |
|
Ceiba 課程網頁 |
http://ceiba.ntu.edu.tw/1041MATH5014_sc |
|
課程簡介影片 |
|
|
核心能力關聯 |
本課程尚未建立核心能力關連 |
|
課程大綱
|
|
為確保您我的權利,請尊重智慧財產權及不得非法影印
|
|
課程概述 |
Optimization (5 weeks)
Basic optimization
Latent variable models and the EM algorithm
Dynamic Programing
Monte Carlo Method (5 weeks)
Monte Carlo integration, importance sampling
Random walk, Markov chain
Metropolis-Hastings algorithm, Gibbs sampler,
Stochastic optimization (5 weeks)
Simulated Annealing
Stochastic gradient descent
Genetic algorithm
Particle swarm
Artificial neural network
|
|
課程目標 |
In this course, we intend to equip students with popular statistical computing tools which are necessary for data analysis, especially in large-scale. In class, we will focus on the core ideas of algorithms. Students will be asked to implement the algorithms on homework exercises after class. Applying algorithms on real data problem is recommended to replace the final exanimation. |
|
課程要求 |
待補 |
|
預期每週課後學習時數 |
|
|
Office Hours |
每週二 10:00~11:30 |
|
指定閱讀 |
待補 |
|
參考書目 |
Andrew Gelman, John B. Carlin, Hal S. Stern, David B. Dunson, Aki Vehtari, and
Donald B. Rubin (2014). Bayesian data analysis (3rd edition)
Christian Robert and George Casella (2009). Introducing Monte Carlo Methods
with R. (2nd edition)
Geoffrey J. McLachlan and Thriyambakam Krishnan (2007). The EM algorithm and
extensions. (2nd edition)
Alexander Shapiro, Darinka Dentcheva, and Andrzej Ruszczynski (2009). Lectures
on Stochastic Programming: Modeling and Theory
http://www2.isye.gatech.edu/people/faculty/Alex_Shapiro/SPbook.pdf |
|
評量方式
(僅供參考) |
|
No. |
項目 |
百分比 |
說明 |
|
1. |
課堂表現 |
10% |
|
2. |
期中考 |
40% |
|
3. |
期末報告 |
50% |
|
|
|
週次 |
日期 |
單元主題 |
|
第1週 |
9/15,9/17 |
Maximum likelihood estimation,
Gaussian mixture model,
Gradient ascent |
|
第2週 |
9/22,9/24 |
Maximum a posteriori estimation,
K-means,
Fuzzy c-means |
|
第3週 |
9/29,10/01 |
EM algorithm |
|
第4週 |
10/06,10/08 |
EM Algorithm
Hidden Markov Model |
|
第5週 |
10/13,10/15 |
Hidden Markov Model
Dynamic Programming |
|
第6週 |
10/20,10/22 |
Monte Carlo Integration |
|
第7週 |
10/27,10/29 |
Importance Sampling |
|
第8週 |
11/03,11/05 |
Random Number Generator |
|
第9週 |
11/10,11/12 |
Markov chain Monte Carlo |
|
第10週 |
11/17,11/19 |
Metropolis-Hastings Algorithm |
|
第11週 |
11/24,11/26 |
自主學習週 |
|
第12週 |
12/01,12/03 |
Gibbs Sampler |
|
第13週 |
12/08,12/10 |
Slice sampling, non-reverersible MCMC |
|
第14週 |
12/15,12/17 |
12/17 期中考 |
|
第15週 |
12/22,12/24 |
Non-reversible MCMC |
|
第16週 |
12/29,12/31 |
Examples |
|
第17週 |
1/05,1/07 |
Final project presentation |
|