Upload
doanlien
View
216
Download
0
Embed Size (px)
Citation preview
Security Level: Confidential
HUAWEI TECHNOLOGIES CO., LTD.
www.huawei.com
Slide title :40-47pt Slide subtitle :26-30pt
Color::white Corporate Font :
FrutigerNext LT Medium Font to be used by customers and
partners : Arial
Computational aspects of machine learning
Gunnar Peters Huawei Sweden
HUAWEI TECHNOLOGIES CO., LTD. Huawei proprietary. No spread without permission. Page 2
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Content
! Huawei Sweden Algorithm Group ! Introduction to learning ! Training neural networks ! Machine Learning in Radio Resource Management
HUAWEI TECHNOLOGIES CO., LTD.
Security Level:
Slide title :40-47pt Slide subtitle :26-30pt
Color::white Corporate Font :
FrutigerNext LT Medium Font to be used by customers and
partners : Arial
www.huawei.com
Huawei Sweden Algorithm Group
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 4
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
The Algorithm Group Huawei Sweden
4GRRMCoordinatedschedulingSmallpacketop5miza5on(XiaojiaLu,CWCOulo)
BasebandReceiverReceiverarchitectureMac/Phycodesign(JinliangHuang,KTH)
5GRRMRRMarchitectureTrafficsteeringMachineLearning(PabloSolda5,KTH)
GunnarPetersTechnicallead
AlgorithmGroupHuaweiSweden
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 5
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Weresorttosimula8ons
Complexity of the system
Resource allocation in Wireless networks " Wedonothaveexactinforma5onofradio
condi5ons
" Trafficloadalwayschanging
" Nostraightforwardclosedformrulemapping
e.g.radiocondi5onstoradioperformance
RRMisanop8malcontrolproblemwheretheunderlyingdynamicsarenotknown
• Simplifiedmodels• Imperfectknowledge
Quan5zedmodula5onandcoding
• Useofsimplifiedcontrolruleswithtunableparameters
• Algorithmsandparametervaluestunedbysimula5ons No precise dynamic model
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 6
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Machine Learning as a solution No precise dynamic model • Reinforcement Learning
• Adaptive optimal control • Model free • Learns the value of different actions through
interaction with the system • Used successfully in robotics, self driving cars
and other Artificial Intelligence applications • The obvious state space for a Radio Access Network
consists of, for example, radio channels, traffic and KPIs for all users in the network.
• We are back in the curse of dimensionality
Too many variables
• Big data • There are efficient methods for automatic extraction of
essential features. • Extremely successful in computer vision and language
processing. • Statistical learning, deep neural networks etc
HUAWEI TECHNOLOGIES CO., LTD.
Security Level:
Slide title :40-47pt Slide subtitle :26-30pt
Color::white Corporate Font :
FrutigerNext LT Medium Font to be used by customers and
partners : Arial
www.huawei.com
Introdcution to learning
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 8
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Classification of machine learning problems ! Unsupervised learning
# Find structure in data # Reduce dimension of data
! Supervised learning # Labeled data # Predict the label of new data
! Reinforcement learning # Learning optimal control by interacting
with a system
?
?
x2
x2
x1
x1
HUAWEI TECHNOLOGIES CO., LTD. Huawei proprietary. No spread without permission. Page 9
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Example of supervised learning
Deep learning uses neural networks to learn hierarchical features on general images. These features simplify the classification.
RecognizinghandwriYendigits
Classifyinganimals
HUAWEI TECHNOLOGIES CO., LTD. Huawei proprietary. No spread without permission. Page 10
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Learning = regression
! Fitting a model in noise
! Have to learn the model order # Low model order -> model error
# High model order -> over fitting
! We need to learn models with good
generalization properties
x xx
x x x
x
x
xx x
xx
x
x
xx x
xx
x
x
y
x
y
x
y
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 11
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Higher dimensions
x1
x2
. ... .. .. ... . .
y
.
This fairly easy in low dimension
! There are good methods for learning model orders
In higher dimensions (n > 500)
! Complex problem
! Number of learning samples needed grow exponentially
with the dimension
In most problem there is a hidden structure in the data
More advanced learning methods
! Neural networks
! Decision trees
! Gaussian processes
Identifies hidden structures and restricts the learning to a low
dimensional manifold (hyper surface)
x
xx x
xx
x
x
y
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 12
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Formulation of classification problems ! Classification is formulated as a
regression of likelihood functions.
! Instead of fitting the data to hard values
(e.g. 0 and 1) we use conditional
probabilities, which are supposed to vary smoothly with x.
x
y =0
=11 xx xxx x x x
x x xx x x0
x
y
1 xx xxx x x x
x x xx x x0
Condi5onalprobabili5es
HUAWEI TECHNOLOGIES CO., LTD.
Security Level:
Slide title :40-47pt Slide subtitle :26-30pt
Color::white Corporate Font :
FrutigerNext LT Medium Font to be used by customers and
partners : Arial
www.huawei.com
Training neural networks
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 14
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Models Linear model
Sigmoid
Generalization to higher dimensions
and model orders
y
1 xx xxx x x x
x x xx x x0
Condi5onalprobabili5es
x0=inputx1
x2x3
x4=input
Weightmatrixbiasoutputvectoratlayeri+1
Neuralnetwork
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 15
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Learning as optimization
Learningsamplesobtainedbyinterac5on
withtherealworld.x(j)andy(j)arevectors.
Lossfunc5on.nnistheapproxima5on
definedbytheneuralnetwork.
Learningnowbecomestheop5miza5onproblem
output input
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 16
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Gradient descent
W
Loss
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 17
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Backprop algorithm for calculating the gradient Forwardpropaga5onofvalues
Backwardpropaga5onofderiva5ves
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 18
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Putting it all together The gradient of the loss function
Can now be calculated using the backprop algorithm
Using e.g. the gradient descent method
We have a computational scheme for learning from training samples
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 19
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Generalization
! We are only training on a subset of, for example, images of animals.
! There is a chance that what we learn only applies to this subset and does not generalize.
! It is not always best to iterate until the loss function is as small as possible. x
xx x
xx
x
x
xx x
xx
x
x
y
TrainingerrorlargeGeneraliza5ongood
TrainingerrorsmallGeneraliza5onpoor
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 20
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Stochastic gradient descent Writing the loss function as a sum over the training samples
The gradient can be written
The stochastic gradient descent means randomly picking training samples and update the
parameters according to
Theory of random iterations now guarantees good generalization
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 21
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Convolutional neural networks
! Still very sensitive to thing like step size and topology of the neural network.
! Test generalization on a subset of available data
Training data Validation data
Numberofitera5ons
Lossfunc5on
TrainingdataValida5ondata
Stopitera5nghere
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 22
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Rprop: Special iteration methods
! Rprop: # Adaptive step size # Different step size for each parameter
# Gradient gives direction but not size of the incremental improvements
IncreaseΔWwhilepar5alderiva5vehassamesignDecreasewhenitchangessign
W
Loss
VerystableDoesnotavoidgeneraliza8onproblem
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 23
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Deep neural networks ! Image data is typically concentrated around low
dimensional manifold (not every 128 by 128 array is an image).
! In deep learning the neural network is split into a lower part
with only a few outputs, and a top part which is problem specific.
! The lower part is generic and reused between different sets of images and different problems. It can therefore be
trained on an ever increasing set of training samples.
! The top part is problem specific, but since the input dimension is low smaller sets of data can be used.. Neuralnetworkusedto
learngeneralfeaturesofimages
Problemspecificneuralnetwork
Likelihoodfunc5ons
~10–20essen5alfeatures
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 24
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Other types of neural networks
! Deep Neural Networks (DNN) # Tries to auto encode the data into a few features
! Convolutional Neural Networks (CNN) # Uses translation invariance of images to reduce the number of weights in
the network.
# Same weights reused on different parts of the image
! Recurrent Neural Networks (RNN) # Feed back loops between the layers are used to introduce memory in the
network
HUAWEI TECHNOLOGIES CO., LTD.
Security Level:
Slide title :40-47pt Slide subtitle :26-30pt
Color::white Corporate Font :
FrutigerNext LT Medium Font to be used by customers and
partners : Arial
www.huawei.com
Machine Learning in Radio Resource Management
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 26
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Huawei Sweden machine Learning software ! The machine learning simulator framework consists of the following components
# Machine Learning $ Learner $ Policy
$ Agent
$ Explorer
# Interface to application $ Feature extractor
! In e.g. the Learner component there will be there will be classes for # Neural Network, # Decision Tree
# Ensemble learners
respectively.
! The Machine Learning part is application independent. The application specific code ´will
reside in the implementation classes of the Feature Extractor component. In this way the
Machine Learning software can be used in any simulator or program with minimal effort.
Learner Policy Explorer Feature extractor
Agent
MachineLearning Applica5oninterface
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 27
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
Highlights of machine learning in RRM ! This framework has been applied to different 3G and 4G use cases
# 3G RoT adaptation $ 25 % capacity gain with no loss in cell edge performance
# HetNet Cell Range Expansion $ Up to 80% gain in hotspot scenarios (where CRE is supposed to improve the performance)
with little loss in other scenarios.
# 4G Uplink power control $ Around 100% gain. This depends on the scenario.
# 4G TX power allocation $ 200 % reduction in TX power
# 4G Single Frequency Network threshold tuning $ 100 – 150 % gain in high load
HUAWEI TECHNOLOGIES CO., LTD. Huawei Proprietary - Restricted Distribution Page 28
Slide title :32-35pt
Color: R153 G0 B0
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Slide text :20-22pt
Bullets level 2-5:
18pt
Color:Black
Corporate Font :
FrutigerNext LT Medium
Font to be used by customers and
partners :
Arial
Top right corner for field-mark, customer or partner logotypes.
---------------- The following nine groups of colors are an example of how our design colors can be used, please take note that you should only use one design color group per slide. For specific usage details, refer to the “Typesetting Standard”.
1. hYp://deeplearning.net/tutorial/
2. DeepLearning,YoshuaBengio,IanGoodfellow,AaronCourville,MITPress.
3. YoshuaBengio,LearningDeepArchitecturesforAI,Founda5onsandTrendsinMachineLearning,2(1),pp.1-127,
2009.
4. YoshuaBengio,AaronCourville,PascalVincent,Representa5onLearning:AReviewandNewPerspec5ves,Arxiv,
2012.
5. JurgenSchmidhuber,DeepLearningandNeuralNetworks:AnOverview,arXiv,2014.
6. LeonBoYou,FrankE.Cur5sy,JorgeNocedalz,Op5miza5onMethodsforLarge-ScaleMachineLearning,arXiv:
1606.04838v1[stat.ML]15Jun2016.
7. MoritzHardt,BenjaminRecht,YoramSinger,Trainfaster,generalizebeYer,stabilityofstochas5cgradient
descent,arXiv.org>cs>arXiv:1509.01240
8. FrancescoDavideCalabrese;EuhannaGhadimi;LeoWang;GunnarPeters,PabloSolda5,LearningRadioResource
Managementin5GNetworks:Framework,Opportuni5esandChallenges,submiYedIEEECommunica5ons
Magazine
References