Bidirectional Long Short-Term Memory (BI-LSTM) - Revision history

BPeat at 03:16, 2 March 2024

2024-03-02T03:16:01Z

BPeat at 02:58, 2 March 2024

2024-03-02T02:58:04Z

BPeat at 01:28, 18 May 2023

2023-05-18T01:28:46Z

BPeat at 17:33, 3 May 2023

2023-05-03T17:33:43Z

BPeat: Text replacement - "http:" to "https:"

2023-03-28T07:26:38Z

Text replacement - "http:" to "https:"

BPeat at 15:15, 19 March 2023

2023-03-19T15:15:50Z

BPeat at 17:44, 17 August 2020

2020-08-17T17:44:28Z

BPeat at 17:27, 11 June 2020

2020-06-11T17:27:51Z

BPeat at 17:15, 11 June 2020

2020-06-11T17:15:48Z

BPeat at 17:01, 11 June 2020

2020-06-11T17:01:11Z

@@ Line 21: / Line 21: @@
 The purpose of the Bi-LSTM is to look at a particular sequences both from front-to-back as well as from back-to-front. In this way, the network creates a [[context]] for each character in the text that depends on both its past as well as its future.
-Bidirectional Long/Short-Term Memory (BiLSTM) look exactly the same as its unidirectional counterpart. The difference is that the network is not just connected to the past, but also to the future. As an example, unidirectional LSTMs might be trained to predict the word “fish” by being fed the letters one by one, where the recurrent connections through time remember the last value. A BiLSTM would also be fed the next letter in the sequence on the backward pass, giving it access to future information. This trains the network to fill in gaps instead of advancing information, so instead of expanding an image on the edge, it could fill a hole in the middle of an image.  Schuster, Mike, and Kuldip K. Paliwal. “Bidirectional recurrent neural networks.” IEEE Transactions on Signal Processing 45.11 (1997): 2673-2681.
+Bidirectional Long/Short-Term [[Memory]] (BiLSTM) look exactly the same as its unidirectional counterpart. The difference is that the network is not just connected to the past, but also to the future. As an example, unidirectional LSTMs might be trained to predict the word “fish” by being fed the letters one by one, where the recurrent connections through time remember the last value. A BiLSTM would also be fed the next letter in the sequence on the backward pass, giving it access to future information. This trains the network to fill in gaps instead of advancing information, so instead of expanding an image on the edge, it could fill a hole in the middle of an image.  Schuster, Mike, and Kuldip K. Paliwal. “Bidirectional recurrent neural networks.” IEEE Transactions on Signal Processing 45.11 (1997): 2673-2681.
 <youtube>fKSijaXq9qM</youtube>
 <youtube>Bp-_DatyUCY</youtube>

@@ Line 8: / Line 8: @@
 [https://www.google.com/search?q=Bidirectional+LSTM+machine+learning+ML+artificial+intelligence ...Google search]
 * [[Recurrent Neural Network (RNN)]] Variants:
 ** [[Long Short-Term Memory (LSTM)]]

@@ Line 18: / Line 18: @@
 * [[Attention]] Mechanism  ...[[Transformer]] ...[[Generative Pre-trained Transformer (GPT)]] ... [[Generative Adversarial Network (GAN)|GAN]] ... [[Bidirectional Encoder Representations from Transformers (BERT)|BERT]]
-The purpose of the Bi-LSTM is to look at a particular sequences both from front-to-back as well as from back-to-front. In this way, the network creates a context for each character in the text that depends on both its past as well as its future.
+The purpose of the Bi-LSTM is to look at a particular sequences both from front-to-back as well as from back-to-front. In this way, the network creates a [[context]] for each character in the text that depends on both its past as well as its future.
 Bidirectional Long/Short-Term Memory (BiLSTM) look exactly the same as its unidirectional counterpart. The difference is that the network is not just connected to the past, but also to the future. As an example, unidirectional LSTMs might be trained to predict the word “fish” by being fed the letters one by one, where the recurrent connections through time remember the last value. A BiLSTM would also be fed the next letter in the sequence on the backward pass, giving it access to future information. This trains the network to fill in gaps instead of advancing information, so instead of expanding an image on the edge, it could fill a hole in the middle of an image.  Schuster, Mike, and Kuldip K. Paliwal. “Bidirectional recurrent neural networks.” IEEE Transactions on Signal Processing 45.11 (1997): 2673-2681.

@@ Line 16: / Line 16: @@
 ** [[Average-Stochastic Gradient Descent (SGD) Weight-Dropped LSTM (AWD-LSTM)]]
 ** [[Hopfield Network (HN)]]
-* [[Attention]] Mechanism  ...[[Transformer]] Model   ...[[Generative Pre-trained Transformer (GPT)]]
+* [[Attention]] Mechanism  ...[[Transformer]] ...[[Generative Pre-trained Transformer (GPT)]] ... [[Generative Adversarial Network (GAN)|GAN]] ... [[Bidirectional Encoder Representations from Transformers (BERT)|BERT]]
 The purpose of the Bi-LSTM is to look at a particular sequences both from front-to-back as well as from back-to-front. In this way, the network creates a context for each character in the text that depends on both its past as well as its future.

@@ Line 5: / Line 5: @@
 |description=Helpful resources for your journey with artificial intelligence; videos, articles, techniques, courses, profiles, and tools
 }}
-[http://www.youtube.com/results?search_query=Bidirectional+LSTM YouTube search...]
+[https://www.youtube.com/results?search_query=Bidirectional+LSTM YouTube search...]
-[http://www.google.com/search?q=Bidirectional+LSTM+machine+learning+ML+artificial+intelligence ...Google search]
+[https://www.google.com/search?q=Bidirectional+LSTM+machine+learning+ML+artificial+intelligence ...Google search]
 * [[Recurrent Neural Network (RNN)]] Variants:

@@ Line 10: / Line 10: @@
 * [[Recurrent Neural Network (RNN)]] Variants:
 ** [[Long Short-Term Memory (LSTM)]]
 ** [[Gated Recurrent Unit (GRU)]]
 ** Bidirectional Long Short-Term Memory (BI-LSTM)