Deep machine learning by Mario Cho

제로부터시작하는인공지능

Artificial Intelligence

중앙대학교의료보안연구소

Mario Cho (조만석)

hephaex@gmail.com

Contents

• 기계학습이란?

• 신경망이란

• 인공지능 SW TensorFlow

• 딥러닝실습

• 딥러닝응용

Mario ChoDevelopment Experience◆ Image Recognition using Neural Network◆ Bio-Medical Data Processing◆ Human Brain Mapping on High Performance

Computing◆ Medical Image Reconstruction

(Computer Tomography) ◆Enterprise System◆Open Source Software Developer

Open Source Software Developer◆ OPNFV (NFV&SDN) & OpenStack◆ Machine Learning (TensorFlow)

Book◆ Unix V6 Kernel Chungan Univercity

Mario Chohephaex@gmail.com

The Future of Jobs

“The Fourth Industrial Revolution, which includes developments in previously disjointed fields such as artificial intelligence & machine-learning, robotics, nanotechnology, 3-D printing, and genetics & biotechnology, will cause widespread disruption not only to business models but also to labor market over the next five years, with enormous change predicted in the skill sets needed to thrive in the new landscape.”

Today’s information

* http://www.cray.com/Assets/Images/urika/edge/analytics-infographic.html

Google (2)

What is the Machine Learning ?• Field of Computer Science that evolved from the study of pattern recognition and computational learning theory into Artificial Intelligence.

• Its goal is to give computers the ability to learn without being explicitly programmed.

• For this purpose, Machine Learning uses mathematical / statistical techniques to construct models from a set of observed data rather than have specific set of instructions entered by the user that define the model for that set of data.

What is Artificial Intelligence?

Artificial IntelligenceUnderstand information, To Learn, To Reason, & Act upon it

Object Recognition

Make predictions on data

Human-Level Object Recognition

• ImageNet• Large-Scale Visual Recognition Challenge� Image Classification / Localization�1.2M labeled images, 1000 classes�Convolutional Neural Networks (CNNs)has been dominating the contest since..� 2012 non-CNN: 26.2% (top-5 error)� 2012: (Hinton, AlexNet)15.3%� 2013: (Clarifai) 11.2%� 2014: (Google, GoogLeNet) 6.7%� 2015: (Google) 4.9%� Beyond human-level performance

Traditional learning vs Deep Machine Learning

Eiffel Tower

Eiffel Tower

RAW data

RAW data

Deep Learning Network

FeatureExtraction

Vectored Classification

Traditional Learning

Deep Learning

What is a neural network?

Yes/No(Mug or not?)

Data (image)

!

x1 ∈!5 , !x2∈!5

x2 = (W1 × x1)+x3 = (W2 × x2)+

x1 x2 x3x4

x5

W4W3W2W1

Neural network vs Learning networkNeural Network Deep Learning Network

Neural Network

W1

W2

W3

f(x)

1.4

-2.5

-0.06

Neural Network

2.7

-8.6

0.002

f(x)

1.4

-2.5

-0.06

x = -0.06×2.7 + 2.5×8.6 + 1.4×0.002 = 21.34

Convolution Feature

Why is Deep Learning taking off?

Engine

Fuel

Large neural networks

Labeled data (x,y pairs)

Training Process

Deep learning - CNN

Deep learning : CNN

The Big Players

Open Source Software for Machine Learning

Caffe

Theano

Convnet.js

Torch7

Chainer

DL4J

TensorFlow

Neon

SANOA

Summingbird

Apache SA

Flink ML

Mahout

Spark MLlib

RapidMiner

Weka

Knife

Scikit-learn

Amazon ML

BigML

DataRobot

FICO

Google prediction API

HPE haven OnDemand

IBM Watson

PurePredictive

Yottamine

Deep Learning

StreamAnalytics

Big DataMachine Learning

Data Mining

Machine Learning As a Service

Pylearn2

Google Tensorflow

* Source: Oriol Vinyals – Research Scientist at Google Brain

Expressing High-Level ML Computations

• Core in C++ • Different front ends for specifying/driving the computation

• Python and C++ today, easy to add more

* Source: Jeff Dean– Research Scientist at Google Brain

Hello World on TensorFlow

Image recognition in Google Map

* Source: Oriol Vinyals – Research Scientist at Google Brain

Deep Learning Hello World == MNIST

MNIST (predict number of image)

CNN (convolution neural network) training

MNIST code

MNIST

Old Character Recognition

Face extraction method

Face recognition data- sets?

Human-Level Face Recognition

• Convolutional neural networks based face recognition system is dominant

• 99.15% face verification accuracy on LFW dataset in DeepID2 (2014)� Beyond human-level recognition

Source: Taigman et al. DeepFace: Closing the Gap to Human-Level Performance in Face Verification, CVPR’14

Image Recognition

* Source: Oriol Vinyals – Research Scientist at Google Brain

Object Classification and Detection

How to the Object recognition ?

Language Generating

* Source: Oriol Vinyals – Research Scientist at Google Brain

Image Caption Generation

Neural Conversational Model

Neuro Painter

Deep Art

Inceptionism

Automatic Colorization of Black and White Images

Image Generate

Image Segmentation

Scene Parsing

[Farabet et al. ICML 2012, PAMI 2013]

Scene Parsing

[Farabet et al. ICML 2012, PAMI 2013]

Auto pilot car

Q-Learning

How do data science techniques scale with amount of data?

GPU

Inspirer Humanity

Thanks you!

Q&A