A Convolutional Attention Network for Extreme Summarization of - - PowerPoint PPT Presentation

a convolutional attention network for extreme
SMART_READER_LITE
LIVE PREVIEW

A Convolutional Attention Network for Extreme Summarization of - - PowerPoint PPT Presentation

A Convolutional Attention Network for Extreme Summarization of Source Code ATTENTION MECHANISM Attention: Withdrawal from some things in order to deal e ff ectively with others ~William James LIBGDX Cross-platform game and


slide-1
SLIDE 1

A Convolutional Attention Network
 for Extreme Summarization of Source Code

slide-2
SLIDE 2
slide-3
SLIDE 3
slide-4
SLIDE 4
slide-5
SLIDE 5
slide-6
SLIDE 6
slide-7
SLIDE 7
slide-8
SLIDE 8
slide-9
SLIDE 9
slide-10
SLIDE 10
slide-11
SLIDE 11
slide-12
SLIDE 12
slide-13
SLIDE 13
slide-14
SLIDE 14
slide-15
SLIDE 15

ATTENTION MECHANISM

➤ “

Attention: Withdrawal from some things in order to deal effectively with others” ~William James

slide-16
SLIDE 16
slide-17
SLIDE 17
slide-18
SLIDE 18
slide-19
SLIDE 19
slide-20
SLIDE 20
slide-21
SLIDE 21
slide-22
SLIDE 22
slide-23
SLIDE 23
slide-24
SLIDE 24
slide-25
SLIDE 25

LIBGDX

➤ Cross-platform game and visualization development

framework

➤ Write code once and use it in multiple platforms ➤ As low level as you want

slide-26
SLIDE 26

GHTORRENT PROJECT

➤ Monitors Github event time lines ➤ Gets the content and their dependencies ➤ Stores raw JSON to MONGODB ➤ Distributed

slide-27
SLIDE 27

POS2VEC

The "Pos2Vec" component is a Deep Belief Network that consists of 4 stacked autoencoders, that are trained layer by layer, unsupervised

slide-28
SLIDE 28

SPEARMINT

Spearmint is a package to perform Bayesian optimization

slide-29
SLIDE 29

➤ Sample weight vectors with the first probability

slide-30
SLIDE 30
slide-31
SLIDE 31
slide-32
SLIDE 32
slide-33
SLIDE 33
slide-34
SLIDE 34

THANK YOU

➤ Sources: ➤ Lectures of Andrew NG ➤ Lectures of Geoffry Hinton

Neural Machine Translation by Jointly Learning to Align and Translate