Latex使用小技巧

1, \iffalse %块注释命令开始....................\fi %块注释命令结束

2, 在画图的时候, subfigure 和 subfig package 不能同时用,否则提示错误:
LaTeX Error: No counter 'subfigure@save' defined.

实验室请了两个法国教授做报告

The first person is the president of the French association for AI andthe second person is a former president.They are visiting our lab until Friday.

Dr. Michele Sebag（http://www.lri.fr/~sebag/）

day & time: 4/8 13:00-14:30

location：W2 3F lecture rooms 5 and 6 (rooms 325 and 326)

title: Data Streaming based on Affinity Propagation; Application to　Autonomic Computing

abstract: Affinity Propagation, a recent clustering algorithm based onmessage propagation techniques, yields quasi-optimal and stablesolutions at the price of a quadratic complexity. Two extensions will bedescribed, one for reducing the complexity and the other for dealingwith online clustering (data streaming). The talk will describe theapplication of StrAP to Autonomic Grid. The EGEE grid system involvesabout 80,000 CPUs, and concurrently processes 300,000 jobs on a 24 hour,7 day basis. The StrAP algorithm provides an online summary of the jobdistribution, amenable to the prompt detection of anomalies.

Marc Schoenauer （http://www.lri.fr/~marc/）

day & time：4/8 14:30-16:00

location：W2 3F lecture rooms 5 and 6 (rooms 325 and 326)

title: Crossing the Chasm for Evolutionary Algorithms

abstract: One drawback of Evolutionary Algorithms, preventing their wideuse outside the Evolutionary Computation community, paradoxically comesfrom their high flexibility. The large number of parameters EAs offer tothe user does indeed make them difficult to fine tune. This talk willsurvey the current state of research in the area of parameter tuning,both off-line and on-line, focusing on the continuous optimization case,where on-line adaptation provides state-of-the-art algorithms.

top rank conference

需要关注的会议:

data mining: KDD, SDM, ICDM

learning: ICML, NIPS, COLT

CVPR

解决matlab中out of memory <转>

这段时间做一个大数据量的仿真，发现经常出现out of memory 的问题，上网搜索了很多办法，如扩大虚拟内存，使用3Gb开关，优化算法流程，使用pack，save, clear函数等。详见http://hi.baidu.com/zwwcqu/blog/item/2ee2261ef7b9eecfa6866943.html 和http://hi.baidu.com/zwwcqu/blog/item/e3b34235c8560f49241f1474.html

但使用后效果均不理想，数据量一旦增加，马上要出现out of memory的问题。后来把程序相关算法优化了一遍，尤其是针对循环嵌套，把没必要的变量clear掉了，把有用的变量先save，后clear掉，最后终于得出较好的结果。

下面介绍一下关于clear、save、load的特殊用法，这对在for或while等多重循环里出现out of memory非常有效。
for k = 1:N % N为循环次数 % ---------------------
var0 = k; % 获得变量var0 %----------------------
string = [sprintf('var_%d', k) ' = var0;' ]; eval(string); % 等价于 var_k = var0; save(sprintf('var_%d.mat', k), sprintf('var_%d')); % 等价于 save var_k.mat var_k
clear(sprintf('var_%d')); % 等价于 clear var_kend 如果要读取刚才存取的变量var_k, (k = 1,2, ..., N).

那么，可以使用如下用法
for k = 1:N
load(sprintf('var_%d.mat', k)); % 等价于 load var_k.mat k = 1,2, ..., N
end

另外，还有一些非常有用的用法。如果用清除刚才读取的变量var_k, k = 1, 2, ..., Nclear '-regexp' '^var_' % 清除所有以“ var_ ”开头的变量还有很多关于save、clear、load等用法，具体help一下。

关于RSRNG ( Robust Supervised Relevance Neural Gas)

Robust Supervised Relevance Neural Gas 是 Supervised Relevance Neural Gas 的增强版本. 主要创新点在于:

(1) 引入 outiler resistance strategy , 使得对随机初始化不敏感,以及使得对一些outiler的出现, 在prototypes 权值更新,

(2) 引入 Bayes Information Criteria, 以确定 prototypes number 的最优数.

(3) 在protypes 上提供的 dissimilarity space 上, 使用 Discriminant Analysis 进行 分类 . (分类效果会比 1-NN 更好)

(4) Relevance Vector 能反映到底哪些特征,在分类中是起到主要作用的.


PS:
(1) 开始做实验的时候,一直怀疑 Robust Supervised Relevance Neural Gas 的稳定性, 认为效果没有Supervised Relevance Neural Gas 上效果好,而且对参数是比较敏感的.

发现 relevance vector 的迭代处理中,有些问题. 自己为了使 relevance vector 每一个元素都为positive, 自己在程序中强加了 abs 操作:
r_vec = abs(r_vec)/norm(r_vec);
其实:
r_vec = r_vec/norm(r_vec); 就可以了，原来结果不稳定的结果 消失了.

而且能在 比较小的 epoches 里面达到稳定.

(2) Robust SRNG 对 训练样本数据中noisy 的情况下, 效果并不如期盼的那样, 所以Robust SRNG 会在训练样本数据没有 noisy data的时候,有优势.

性能比较:
(1) KNN的性能:
th_hld: 0.420000 : Detection Rate: 0.828571 , False Positive: 0.200000 th_hld: 0.440000 : Detection Rate: 0.885714 , False Positive: 0.200000 th_hld: 0.460000 : Detection Rate: 1.000000 , False Positive: 0.205714
运行时间: 1.78秒

(2) Kernel Binary Weight - KNN:
th_hld: 0.100000 : Detection Rate: 0.771429 , False Positive: 0.071429 th_hld: 0.120000 : Detection Rate: 0.828571 , False Positive: 0.085714 th_hld: 0.140000 : Detection Rate: 1.000000 , False Positive: 0.094286
运行时间: 3.62 秒 (运行时间主要花在 Binary Weight Matrix的计算上)

(3) Robust SRNG:
the number of prototypes: 20
AIC: 3.977546 , BIC: 4.275054
1nn : Detection Rate: 0.914286 , False Positive: 0.031429
fld : Detection Rate: 1.000000 , False Positive: 0.028571
运行时间: 2.63秒

可以看出在dissimilarity space 用FLD来分类,效果比1-NN好,

从运行时间上来 T(KNN) < T(RSNG) < T(KBWKNN)
效果在里面是最好的.

Conf. Info

PKDD2009:
http://www.ecmlpkdd2009.net/information/key-dates/
April 20th

ICDM2009:
http://www.cs.umbc.edu/ICDM09/
June 26

ICDE2010:
http://www.icde2010.org/
Coming soon.

ICCCAS09:
http://icccas09.uestc.edu.cn
Feb 28, 2009

JCST ,Special issue on Machine Learning
http://jcst.ict.ac.cn/tempfile/JCST_AMLA.htm
May 01, 2009

KNN 和 KBWC-KNN 在 Clean Data 和 Noisy Data的性能

Fig .1 KNN 和 KBWC-KNN 在 Clean Data 和 Noisy Data的性能 (50 维数)

也正如 论文提到的那样:

KBWC-KNN 比 KNN 具有 capability of toleration of noisty training data.

自己设计的RSRNG(Robust Supervised Neural Gas) is supposed to make better performance than KBWC-KNN in both clean training data and noisy training data circumstances.

references:

(1) [06'] Intrusion Detection using Text Processing Techniques with a Binary-Weighted Cosine Metric

(2) [02'] Using Text Categorization Techniques for Intrusion Detection

(3) [07'] Intrusion detection using text processing techniques with a kernel based similarity measure.pdf