11. Feature Vector Representation
Bengio, 2014, Representation Learning: A Review and New Perspectives
FeatureVector
Representation
Learning
http://deeplearning4j.org/convolutionalnets
NN
feature
extractor
15. Sentence Retrieval
Results:
• Training acc: 99% up
• Validation acc: 99% up
Experiment:
• 5 classes sentences
• Training Set 80%
• Validation Set 20%
Feature vector PCA tSNE
16. Sentence Retrieval
Recurrent Neural Network
Kyunghyun Cho et. al., 2014,
Learning Phrase Representations using
RNN Encoder–Decoder for Statistical
Machine Translation
https://devblogs.nvidia.com/parallelforall/
introduction-neural-machine-translation-gpus-part-2/
17. Skip-Thought Vectors
Skip-Thought Vectors http://arxiv.org/pdf/1506.06726v1.pdf
Encode a sentence into a thought vector hi by predicting its neighbor
sentence.
Ryan Kiros et. al., 2015, Skip-Thought Vectors
… I just got back home. I could see the cat on the steps. This was strange. …
21. Limitations
Skip-Thought Vectors
• Requiring huge size of corpus
• … I just got back home. I could see the cat on the steps. This was strange. …
• … I got back to office. I could see the cat on the steps. This was cool. …
11,051 novels with 17,515,150 sentences
• Scenario dependency
query sen skt result skt score
0.697
… … 0.697
0.697
24. Image Retrieval
Input
Image
pre-
processing
(2)
(1)
Hashing
Vector
Layer
FC + Softmax
Layer
.
.
.
.
.
.
CNN
network
1st: various targets
Step1
Training
copy
CNN
network
Step2
Training
Kevin Lin et. al, 2015, Deep Learning of Binary Hash Codes for Fast Image Retrieval
.
.
.
.
.
.
2nd: task specific targets
encoding
application Hashing
Vector
Layer
copy
CNN
network
copy
Feature Vector
AlexNet
VGG
25. Image Retrieval
Kevin Lin et. al, 2015, Deep Learning of Binary Hash Codes for Fast Image Retrieval
Demo
http://tweddielin.pythonanywhere.com/
https://github.com/tweddielin/flask-imsearch
Github
26. Conclusion
1. Deep Learning => Trainable Feature Extractor
2. Type of Neural Nets: MLP, CNN, RNN
3.
• http://www.slideshare.net/tw_dsconf/ss-62245351
• http://speech.ee.ntu.edu.tw/~tlkagk/courses_MLSD15_2.html