1.
Deep Learning for
Recommender Systems
Alexandros Karatzoglou
Senior Research Scientist @ Telefonica Research
first.lastname@gmail.com
@alexk_z

2.
Telefonica Research
Machine Learning
HCI
Network & Systems
Mobile Computing
http://www.tid.es

3.
Why Deep?
ImageNet challenge error rates (red line = human performance)

4.
Why Deep?

5.
Inspiration for Neural Learning
Early aviation
attempts aimed at
imitating birds, bats

6.
Neural Model

7.
Neuron a.k.a. Unit

8.
Feedforward Multilayered Network

9.
Learning

10.
Stochastic Gradient Descent
Generalization of (Stochastic) Gradient Descent

11.
Stochastic Gradient Descent

12.
Stochastic Gradient Descent

13.
Stochastic Gradient Descent

14.
Feedforward Multilayered Network

15.
Backpropagation

16.
Backpropagation
Does not work well in plain a “normal”
multilayer deep network
Vanishing Gradients
Slow Learning
SVM’s easier to train
2nd Neural Winter

17.
Modern Deep Networks
Ingredients:
Rectified Linear Activation function a.k.a. ReLu

18.
Modern Deep Networks
Ingredients:
Dropout:

19.
Modern Deep Networks
Ingredients:
Mini-batches:
Stochastic Gradient Descent
Compute gradient over many (50 -100)
data points (minibatch) and update.

20.
Modern Deep Networks
Ingredients:
Softmax output:

21.
Modern Deep Networks
Ingredients:
Categorical cross entropy loss

22.
Modern Feedforward Networks
Ingredients:
Batch Normalization

23.
Modern Feedforward Networks
Ingredients:
Adagrad a.k.a. adaptive learning rates

24.
Restricted Boltzmann Machines

25.
Restricted Boltzmann Machines

26.
Convolutional Networks

27.
Convolutional Networks
[Krizhevsky 2012]

28.
Convolutional Networks
[Faster R-CNN: Ren, He, Girshick, Sun 2015] [Farabet et al., 2012]

29.
Convolutional Networks
[Faster R-CNN: Ren, He, Girshick, Sun 2015] [Farabet et al., 2012]

30.
Convolutional Networks
Self Driving Cars
Convolutional example slides from Fei-Fei Li & Andrej Karpathy & Justin Johnson Lecture 6 75

31.
Convolutional Networks
Standford CS231n: Convolutional Neural Networks for Visual Recognition

32.
Convolutional Networks

33.
Convolutional Networks

34.
Convolutional Networks

35.
Convolutional Networks

36.
Convolutional Networks

37.
Convolutional Networks

38.
Convolutional Networks
AlexNet [Krizhevsky et al 2014]

39.
d
d
D-tour → Matrix Factorization
2
5
4
5
4
4
1
5
5
4
1
2
5
2
4
1

40.
Convolutional Networks for
enhancing Collaborative Filtering
VBPR: Visual Bayesian Personalized Ranking from Implicit Feedback He,
etl AAAI 2015

41.
Convolutional Networks for Music
feature extraction
Deep Learning can be used to learn item profiles e.g. music
Map audio to lower dimensional space where it can be used
directly for recommendation
Useful in recommending music from the long tail (not popular)
A solution to the cold start problem

42.
Convolutional Networks for Music
feature extraction
A. van den Oord, S. Dielmann, B. Schrauwen Deep content-
based music recommendation NIPS 2014

43.
Convolutional Networks
deepart.io

44.
Recurrent Neural Networks

45.
Recurrent Neural Networks
Long Short Term Memory

46.
Recurrent Neural Networks

47.
Recurrent Neural Networks
PANDARUS:
Alas, I think he shall be come approached and the day
When little srain would be attain'd into being never fed,
And who is but a chain and subjects of his death,
I should not sleep.
Second Senator:
They are away this miseries, produced upon my soul,
Breaking and strongly should be buried, when I perish
The earth and thoughts of many states.
DUKE VINCENTIO:
Well, your wit is in the care of side and that.
Second Lord:
They would be ruled after this chamber, and
my fair nues begun out of the fact, to be conveyed,
Whose noble souls I'll have the heart of the wars.
Clown:
Come, sir, I will make did behold your worship.
VIOLA:
I'll drink it.

48.
Recurrent Neural Networks

49.
Recurrent Neural Networks

50.
Recurrent Neural Networks

51.
Recurrent Neural Networks

52.
Session-based recommendation
with Recurrent Neural Networks
RNN (GRU) with ranking loss function
ICLR 2016 [B. Hidasi, et.al.]
Treat each user session as sequence of clicks

53.
Session-based recommendation
with Recurrent Neural Networks
RNN (GRU) with ranking loss function
ICLR 2016 [B. Hidasi, et.al.]
Treat each user session as sequence of clicks

54.
Autoencoders

55.
Autoencoders

56.
Autoencoders

57.
Personalized Autoencoders
Collaborative Denoising Auto-Encoders for Top-N Recommender Systems Wu
et.al. WSDM 2016

58.
(Some) Deep Learning Software
Theano: Python Library
TensorFlow: Python Library
Keras: High Level Python Library (Theano &TF)
MXNET: R, Python, Julia

59.
Thanks
● Some slides or parts of slides are taken from
other excellent talks and papers on Deep
Learning (e.g. Yan Lecun, Andrej Karpathy and
other great deep learning researchers)