Zhiyuan Thang 15-06-29 - 版本历史

2015年6月30日 (二) 14:47 Tangzy

2015-06-30T14:47:54Z

2015年6月30日 (二) 14:46 Tangzy

2015-06-30T14:46:36Z

2015年6月30日 (二) 14:36 Tangzy

2015-06-30T14:36:32Z

Tangzy：以“ Last few weeks (lazy): Presently got the conclusion that randoming the weights ahead of softmax of DNN before MPE training may be not helpful. Using language vec...”为内容创建页面

2015-06-30T14:31:13Z

以“ Last few weeks (lazy): Presently got the conclusion that randoming the weights ahead of softmax of DNN before MPE training may be not helpful. Using language vec...”为内容创建页面

新页面

Last few weeks (lazy):

Presently got the conclusion that randoming the weights ahead of softmax of DNN before MPE training may be not helpful.

Using language vector to pre-train the hidden layers of DNN gives at least a little improvement.

have not realized leaky Rectifier and temporal Rectifier, CUDA programing may be needed.

This week:

realize the two kinds of Rectifier.

@@ 第16行： / 第16行： @@
 This week:
-realize the two kinds of Rectifier;
+realize the two kinds of Rectifiers;
 paper reading.

←上一版本		2015年6月30日 (二) 14:46的版本
第1行：		第1行：

	Last few weeks (lazy):		Last few weeks (lazy):
		+
		+	training LSTM with MPE (failed);

	presently got the conclusion that randoming the weights ahead of softmax of DNN before MPE training may be not helpful;		presently got the conclusion that randoming the weights ahead of softmax of DNN before MPE training may be not helpful;

@@ 第2行： / 第2行： @@
 Last few weeks (lazy):
-Presently got the conclusion that randoming the weights ahead of softmax of DNN before MPE training may be not helpful.
+presently got the conclusion that randoming the weights ahead of softmax of DNN before MPE training may be not helpful;
-Using language vector to pre-train the hidden layers of DNN gives at least a little improvement.
+using language vector to pre-train the hidden layers of DNN gives at least a little improvement, especially 3 hidden layers is pre-trained when DNN is of 4 hidden layers;
 This week:
-realize the two kinds of Rectifier.
+realize the two kinds of Rectifier;