File:2017 RecurrentHighwayNetworks Fig1a.png
Jump to navigation
Jump to search
2017_RecurrentHighwayNetworks_Fig1a.png (364 × 389 pixels, file size: 28 KB, MIME type: image/png)
Summary
Figure 1: Comparison of (a) stacked RNN with depth [math]\displaystyle{ d }[/math] and (b) Deep Transition RNN of recurrence depth [math]\displaystyle{ d }[/math], both operating on a sequence of [math]\displaystyle{ T }[/math] time steps. The longest credit assignment path between hidden states [math]\displaystyle{ T }[/math] time steps is [math]\displaystyle{ d \times T }[/math] for Deep Transition RNNs. In: Zilly et al. (2017)
File history
Click on a date/time to view the file as it appeared at that time.
Date/Time | Thumbnail | Dimensions | User | Comment | |
---|---|---|---|---|---|
current | 09:03, 4 August 2019 | 364 × 389 (28 KB) | Omoreira (talk | contribs) | '''Figure 1:''' Comparison of (a) stacked RNN with depth <math>d</math> and (b) Deep Transition RNN of recurrence depth <math>d</math>, both operating on a sequence of <math>T</math> time steps. The longest [[credit ass... |
You cannot overwrite this file.
File usage
The following page uses this file: