Xingchao work - 版本历史

2016年4月8日 (五) 04:44 Xingchao

2016-04-08T04:44:32Z

Xingchao：以“=Chaos Work= ==SLT==”替换内容

2016-04-08T04:43:50Z

以“=Chaos Work= ==SLT==”替换内容

2015年10月14日 (三) 10:59 Xingchao

2015-10-14T10:59:57Z

Xingchao：/* Date Line */

2015-10-12T06:42:28Z

‎Date Line

Xingchao：/* Date Line */

2015-10-09T10:04:23Z

‎Date Line

Xingchao：/* Date 2015-10-09 */

2015-10-09T10:02:42Z

‎Date 2015-10-09

2015年10月8日 (四) 10:05 Xingchao

2015-10-08T10:05:19Z

2015年10月8日 (四) 10:04 Xingchao

2015-10-08T10:04:07Z

Xingchao：/* Reproduce Nested Dropout */

2015-07-02T11:25:44Z

‎Reproduce Nested Dropout

Xingchao：/* =Simple semi-linear autoencoder */

2015-07-02T11:25:19Z

‎=Simple semi-linear autoencoder

←上一版本		2016年4月8日 (五) 04:44的版本
第1行：		第1行：
	=Chaos Work=		=Chaos Work=
−	==SLT==	+	[[SLT]]

@@ 第1行： / 第1行： @@
 =Chaos Work=
+==SLT==
-==Date Line==
-−
-===Date 2015-10-09===
-−
-====Plan to do====
-−
-:1. Write Matrix Factorization program.
-:2. Write DNN Max-Margin program with dropout.
-:3. Run some test model in GPU machines.
- <Written by 10-08>
-====Done====
-:2. Write DNN Max-Margin program with dropout.
-:3. Run some test model in GPU machines.
-−
-−
-===Date 2015-10-10===
-−
-====Plan to do====
-−
-:1. Write Matrix Factorization program.
-:2. Start test different matrix factorization lost functions.
-:3. Write Binary Vector for theano mode.
- <Written by 10-09>
-====Done====
-:1. Write Matrix Factorization program.
-:2. Start test different matrix factorization lost functions.
-<Written by 10-12>
-−
-===Date 2015-10-12===
-−
-====Plan to do====
-−
-:1. Test different matrix factorization lost functions.
-:2. Write Binary Vector for theano mode.
-:3. Thinking the method for merge different resource for one matrix factorization process.
-====Done====
-:1. Test different matrix factorization lost functions.
-:2. Write Binary Vector for theano mode.
-:3. Thinking the method for merge different resource for one matrix factorization process.
- <Written by 10-13>
-−
-===Date 2015-10-13===
-====Plan to do====
-:1. Read Papers for matrix factorization.
-:2. Read Papers for binary vectors.
-====Done====
-:1. Read Papers for matrix factorization.
-    Paper list:
-      Neural Word Embedding as Implicit Matrix Factorization. Omer Levy;
-:2. Read Papers for binary vectors.
-    Paper list:
-      Learning to Hash for Indexing Big Data - A Survey. Jun Wang;
-      Learning Hash Codes with Listwise Supervision. Jun Wang;
-      Kernelized Locality-Sensitive Hashing for Scalable Image Search. Brian Kulis;
- <Written by 10-14>
-−
-===Date 2015-10-13===
-====Plan to do====
-:1. Read Papers for matrix factorization.
-:2. Read Papers for binary vectors.
-====Done====

@@ 第36行： / 第36行： @@
 :2. Write Binary Vector for theano mode.
 :3. Thinking the method for merge different resource for one matrix factorization process.
-  <Written by 10-12>
+====Done====
 ====Done====

@@ 第24行： / 第24行： @@
 :3. Write Binary Vector for theano mode.
   <Written by 10-09>
 ====Done====

@@ 第12行： / 第12行： @@
   <Written by 10-08>
 ====Done====
-:1. Write Matrix Factorization program.
+:2. Write DNN Max-Margin program with dropout.
 :3. Run some test model in GPU machines.

←上一版本		2015年10月9日 (五) 10:02的版本
第11行：		第11行：
	:3. Run some test model in GPU machines.		:3. Run some test model in GPU machines.
	<Written by 10-08>		<Written by 10-08>
		+	====Done====
		+	:1. Write Matrix Factorization program.
		+	:3. Run some test model in GPU machines.

@@ 第3行： / 第3行： @@
 ===Reproduce Nested Dropout===
 Nested dropout method proposed by Rippel et. in their paper "Learning Ordered Representations with Nested Dropout", they proposed a dropout method which could learning ordered information in different dimensions.
 ====Simple semi-linear autoencoder====
 Their first draft work is semi-linear autoencoder, so I will reproduce this work.