- Week 6 – Practicum: RNN and LSTM architectures

Week 6 – Practicum: RNN and LSTM architectures

Course website: http://bit.ly/DLSP20-web
Playlist: http://bit.ly/pDL-YouTube
Speaker: Alfredo Canziani
Week 6: http://bit.ly/DLSP20-06

0:00:00 – Week 6 – Practicum

PRACTICUM: http://bit.ly/DLSP20-06-3
We discussed architecture of Vanilla RNN and LSTM models and compared the performance between ...
Course website: http://bit.ly/DLSP20-web
Playlist: http://bit.ly/pDL-YouTube
Speaker: Alfredo Canziani
Week 6: http://bit.ly/DLSP20-06

0:00:00 – Week 6 – Practicum

PRACTICUM: http://bit.ly/DLSP20-06-3
We discussed architecture of Vanilla RNN and LSTM models and compared the performance between the two. LSTM inherits advantages of RNN, while improving RNN’s weaknesses by including a ‘memory cell’ to store information in memory for long periods of time. LSTM models significantly outperforms RNN models.
0:00:21 – Overview and 4 types of RNN
0:14:15 – RNN Training and related issues
0:30:38 – Long Short-Term Memory and Notebook Examples

#Deep Learning #Yann LeCun #RNN #LSTM #Recurrent Neural Network #Sequence #Jürgen Schmidhuber

スペースでAND検索・ORで区切るとOR検索・先頭に「-」で除外検索

– Week 6 – Practicum - Week 6 – Practicum: RNN and LSTM architectures

– Week 6 – Practicum

Week 6 – Practicum: RNN and LSTM architectures
2020年04月13日　
00:00:00 - 00:00:21
– Overview and 4 types of RNN - Week 6 – Practicum: RNN and LSTM architectures

– Overview and 4 types of RNN

Week 6 – Practicum: RNN and LSTM architectures
2020年04月13日　
00:00:21 - 00:14:15
– RNN Training and related issues - Week 6 – Practicum: RNN and LSTM architectures

– RNN Training and related issues

Week 6 – Practicum: RNN and LSTM architectures
2020年04月13日　
00:14:15 - 00:30:38
@ with respect to the example which you were talking about sequence predictionLet's say at test time, if there is small mistake in one of the time step, will it carried across rest of the time steps or handled in a better way by hidden state ?Ex : Instead of predicting "ciou" at time step 2 if RNN outputs "cipu" will this mistake gets carried over to the rest of the time steps ? - Week 6 – Practicum: RNN and LSTM architectures

@ with respect to the example which you were talking about sequence predictionLet's say at test time, if there is small mistake in one of the time step, will it carried across rest of the time steps or handled in a better way by hidden state ?Ex : Instead of predicting "ciou" at time step 2 if RNN outputs "cipu" will this mistake gets carried over to the rest of the time steps ?

Week 6 – Practicum: RNN and LSTM architectures
2020年04月13日　
00:20:20 - 00:53:34
At u said that the shown example is kind of 4 RNNs, what did u mean by this? I can't find more than 2 on the slide 😀 - Week 6 – Practicum: RNN and LSTM architectures

At u said that the shown example is kind of 4 RNNs, what did u mean by this? I can't find more than 2 on the slide 😀

Week 6 – Practicum: RNN and LSTM architectures
2020年04月13日　
00:30:24 - 00:53:34
– Long Short-Term Memory and Notebook Examples - Week 6 – Practicum: RNN and LSTM architectures

– Long Short-Term Memory and Notebook Examples

Week 6 – Practicum: RNN and LSTM architectures
2020年04月13日　
00:30:38 - 00:53:34
@Dmitry Д just count the weight matrices W at . You'll see, you have 4 W matrices. Those are the 4 RNN I'm talking about. - Week 6 – Practicum: RNN and LSTM architectures

@Dmitry Д just count the weight matrices W at . You'll see, you have 4 W matrices. Those are the 4 RNN I'm talking about.

Week 6 – Practicum: RNN and LSTM architectures
2020年04月13日　
00:31:31 - 00:53:34
チャンネル登録YouTube RSS

Alfredo Canziani

※本サイトに掲載されているチャンネル情報や動画情報はYouTube公式のAPIを使って取得・表示しています。

概要 カレンダー 動画一覧 タイムテーブル チャンネル分析

Timetable

動画タイムテーブル

スペースでAND検索・ORで区切るとOR検索・先頭に「-」で除外検索

動画数：129件

もっと見る