SlideShare a Scribd company logo
1 of 59
Deep Learning: Towards General
Artificial Intelligence
Dr. Rukshan Batuwita (Machine Learning Scientist)
Senior Data Scientist
Ambiata Pvt Ltd, Sydney, Australia
What is Artificial Intelligence?
• Field of Study to Develop Machines that act
like Humans!
• Recent Definition: Develop Machines that act
rationally!
General AI
• Learning and Reasoning
• Planning
• Adaptability
• Vision
• Speech recognition
• Automation
• Mobility
• etc.
Narrow AI
• Learning and Reasoning
• Planning
• Adaptability
• Vision
• Speech recognition
• Automation
• Mobility
• etc.
• People have been working on for the last 50 years
• This has many applications
Narrow
AI
What is Machine Learning?
• Computer Program: Input to output mapping
Computer Program
(Algorithm/List of
Instructions)
Inputs Outputs
When we know the algorithm to solve a task, then we can program it
?Inputs Outputs
AI Problems:
Ex.
? Cat or Dog
In Machine Learning…
Algorithm
(Model) Cat or Dog
Machine Learning techniques
. . .
. . .
Train/Learn
Introduction to
Artificial Neural
Networks
(Biologically Inspired)
Biological Neurons
Inputs Outputs
Biological Brain Biological Neural Network
Biological Neuron
Processing/Computing
Biological Learning
• Biological Neuron
Learning happens due to some chemical reactions in synaptic connections
A synaptic connection
A typical adult human brain has
about 1014 synapses (connections)
Artificial Neuron
• A computational model
Y
y = f ( wj xj
j=1
d
å )
• Called ‘Perceptron’
• Introduced in 1960’s
• Weights can be learned by an optimization method
like Gradient Descent
Weights (represent
chemicals)
Inputs
Outputs
Processing
/Computing
Inputs
Inputs
Perceptron
Perceptron = Linear Regression
Perceptron = Logistic Regression
Activation Functions:
Y
y = f ( wj xj
j=1
d
å )
Artificial Neural Network
weights
weights
• Artificial Neurons are corrected together to form a network
• Called Multi Layer Perceptron (MLP)
• A Non-linear model of the parameters
• Trained by popular Backpropagation (Gradient Descent)
Backpropagation – Main Idea
1. Calculate Error/Loss = f(Label , Prediction)
2. Calculate Gradient/Derivative of the Loss w.r.t. each weight
3. In order to calculate the gradient of the inner weights,
apply the chain rule of derivatives
4. Update each weight in the direction of the negative gradient (Gradient
Descent)
Error = f(label, prediction)
Evolution of Neural
Networks
• Quite popular in 1980’s and 1990’s
• Worked well for some pattern recognition
problems:
– Ex: Handwritten digit recognition Le-Net
used by US postal department
• Other ML methods (ex. Kernel methods such as
SVMs) dominated ANNs in early 2000’s
• Main problems of ANNs:
– Local-minima (since the loss function is non-convex)
– Difficult to train networks with more then 3/4 layers
• Overfitting
• Computational time
• Vanishing Gradient problem (e.g. when Sigmoid
activation is used)
• (didn’t work well in more complex problems like general
image classification)
Before 2006…
(LeCun et al., 1998)
Yan LeCun, NYU
Geoff Hinton,
Uni Torento
Yoshua Bengio,
Uni Montreal
After 2006…
• Several major breakthroughs happened giving birth to
Deep Learning
• In general, Deep Learning is nothing but good old
Neural Networks with many layers:
…
N
…
• Deep Learning methods have been significantly
outperforming the existing methods in major Computer
Vision and Speech Recognition competitions since 2010
ImageNet Results…
About 14M images of 100k categories/concepts
Main Advancements
made Deep Learning
possible
1. Unsupervised Feature Learning
• In classical Machine Learning:
Feature ExtractionRaw Data Feature pre-
processing
Model Learning
80%-90% of the effort (Human effort)
• In Deep Learning:
Feature LearningRaw Data Model Learning
Deep Learning
+
Model
Model
Feature Learning = Representation Learning = Embedding Learning
Feature Learning/Representation Learning
(Ex. Face Detection)
Layer 1
(Detects Edges)
Layer 2
(Detects Face parts
Combination of edges)
Deeper layer
(Detects Faces)
Input
Pixels
InputPixels
Techniques for Representation
Learning
1. Layer-wise unsupervised pre-training
1. Stacked Autoencoders
Input Output
Encode Decode
Edge Detectors
Autoencoder
• No labels required
• Unsupervised Training
Pixelinput
Pixeloutput
Stacked Autoencoders
1. Train one layer autoencoder at a time [unsupervised learning] and stack
them
2. Then train the final network using the available labels [supervised learning]
Low level features
Higher level features
Higher level features
INPUT LABEL
Techniques for Representation
Learning
Input
1. Layer-wise unsupervised pre-training
2. Deep Belief Networks (Restricted Boltzmann
MMachines (RBM) are stacked together)
Techniques for Representation
Learning
Techniques for Representation
Learning
2. Deep Convolution Networks
Convolution Filters
Kernel/convolution matrix/mask/filter
Edge Detector
X_1 … …
… … …
… … X_9
W_1 … …
… … …
… … W_9
zi = xiwi
i=1
9
å
X
3x3 Image patch
Z
CONV( ),
Techniques for feature learning
2. Deep Convolution Networks
Feature Extraction Classification
• Convolutional Filters (low-level and high-level) are also learned automatically with Backprop
Subsampling = average, max (max pooling) - noise reduction
Different types of filters result in
different feature maps
Techniques for feature learning
2. Deep Convolution Networks
Inputlayer
W_1
x1
x1 x2
x3 x4
W_2
W_3
W_4
x5
x6
x2
x3
x4
x5
x6
W_1
W_2
W_3
W_4
… … …
…
…
…
W_1 W_2
W_3 W_4
2X2 filter
x5
x6
… … …
…
…
…
W_1 W_2
W_3 W_4
x1
x3
… … …
…
…
…
W_1 W_2
W_3 W_4
• Each layer is represented by connected neurons
• Each convolution layer is connected to the previous layers sparsely and with shared weights
Techniques for feature learning
2. Deep Convolution Networks
• Convolution and Subsampling (Pooling) leads to detect translational invariance features
• Works with language (document classification, translation) and Voice recognition
Motivations for
Feature/Representation/Embeddin
g Learning
Motivations for
Feature/Representation learning
1. Cut down the effort of handcrafting features
2. Hierarchical, distributed, compositional knowledge
representations in Brain
– Humans organize their concepts and ideas hierarchically
– Humans first learn simple concepts and compose them
together to represent complex ideas
– Human problem solving/Engineering (multiple level of
abstractions)
– Human language understanding
– Pattern recognition in brain, etc.
Motivations for
Feature/Representation learning
• Hierarchical, distributed, compositional
knowledge representation/pattern recognition
in Brain
Pattern Recognition in Brain Pattern Recognition
In Deep Learning
Motivations for
Feature/Representation learning
3. Power of distributed, compositional
representations
• Concepts are represented as composition of features
at different levels
• The number of concepts can be represented grow
exponentially with the size of the network
Input
Low-level representations (e.x. edges) High-level representations
Motivations for
Feature/Representation learning
4. Manifold Learning
• Assumption: Input data has some structure (not 100%
random) which is concentrated in a lower-dimensional
manifold of the original features
• Ex: most of the arbitrary pixel value configurations don’t create
the images of faces
• Representation in each layer can be considered as a
learned manifold of the previous layer
28!
F or AI T a sk s: Ma n i f ol d st r uct ur e
• examples!concentrate!near!a!lower!dimensional!“manifold!
• Evidence:$most$input$configuraDons$are$unlikely$
Pixels (32*32 image)
E.x.
Motivations for
Feature/Representation learning
5. Transfer Learning
– Generalization: ability of a model to predict well on
unseen test data
– Representation of complex concepts -> Deep
Networks
– Good generalization of complex models like Deep
Neural Networks rely on the availability of large
number of labeled training data
– Most of the available data are not labeled
– In Transfer Learning
1. Train a Deep Network with unlabeled data in unsupervised
manner
2. Use the available labeled data to train the required model
Motivations for
Feature/Representation learning
5. Transfer Learning
Example: Image recognition model
. . .
Unsupervised
pre-training with unlabeled data
to learn the representations of
different levels of abstraction
Transfer the
knowledge
car
Supervised Learning with
available labeled data
...
Hu
man
Variations of Transfer Learning
• Multi-Instance Learning (when labels are not
available at the instance level)
Document Classification Model
Based on the similarity of the sentence/word
embedding [Kotzias, Denil and deFreitas, 2014]
Variations of Transfer Learning
• Max-margin Learning without labels
[From machine learning to machine reasoning, Leon Bottou, 2014]
Variations of Transfer Learning
• Max-margin Learning without labels
[NLP almost from scratch, Ronan Collobert et al., 2011]
Other advancements
made Deep Learning
possible
Other advancements…
• ‘Dropouts’ regularization for training with
Backpropagation for higher generalization
• Rectified Linear Functions instead of Sigmoid
(avoid vanishing gradient problem)
Other architectures…
• Memory Networks (LSTM)
– Question answering
• Recurrent Networks
– Detecting inputs with sequential relationships
(voice recognition)
• Combination of existing architectures
Improved Computing Power…
GPU Computation
– Parallel Neural Network Training on GPU clusters
(ideal for simple Matrix/Vector operations, hence for
backpropagation)
– Reduced the training time of deep networks from
weeks to days
– NVIDIA CUDA Deep Neural Network library
Improved Computing Power…
• Commodity Hardware
– Multi-core single machines, clusters, GUP clusters
• Open source software
– Torch (open source ML library,
https://github.com/torch/torch7/)
– From Yoshua Bengio’s group
http://deeplearning.net/software/theano/)
– Caffe
– Google TensorFlow
Industrial Applications of
Deep Learning
Techniques
Google Brain Project
– Started by Andrew Ng in 2011
– In 2012: Neural Network with 1 Billion connection
was trained across 16,000 CPU cores
– They considered this ANN as simulating a very small-
scale “newborn brain,” and show it YouTube video
for a week, what will it learn?
– Used an Unsupervised (Self-taught-learning) to learn
features from unlabeled Google images –
Autoencoder
– Exposed to fames of 10M YouTube videos over a
week
http://googleblog.blogspot.com.au/2012/06/using-large-scale-brain-simulations-for.html
http://static.googleusercontent.com/media/research.google.com/en//archive/unsupervised_icml2012.pdf
Andrew Ng,
Standford
Google Brain Project
What Happened?
• One of the artificial neurons learned to respond strongly
to pictures of Cats.
Evolution of Deep Leaning at Google
– Google has been heavily investing on Deep
Learning research
– In 2013 Google hired Geoff Hinton and acquired
his start-up company DNNResearch Inc.
– In 2014 they purchased a UK-based Machine
Learning company called DeepMind Technologies
for estimated $650 Million
Deep Mind
Apollo Program for AI
Working towards solving General AI with
Deep Reinforcement Learning….
DeepMind
• Famous paper: Applying Deep RL to train agents to play classic Atari games
DeepMind Video
• https://www.youtube.com/watch?v=V1eYniJ0
Rnk
AlphaGo
• Traditional Chinese game - Go
• The most complex board game of all
• Alpha Go beat the world champion in Go 4/5
Lee Sedol
Deep Dream
(http://deepdreamgenerator.com)
• What features will be picked up by Google’s
Deep ANNs?
Deep ANN
Original
Image
Original
Image + Recognized
Features
Google Voice Recognition (in Android and Search by Voice)
Deep Learning Products at Google
Google search by Image
(Search for similar images to an uploaded
image)
Facebook
• Yann Lecun is the head of Facebook AI Research
• Face Recognition: Deep Face
• claim to have close to human-level performance
• Personal Assistant:
Facebook M
Other…
• Microsoft Cortana, Skype Translate
• Nvida Self Driving Cars
• Image Captioning Systems
• Siemens Medical Image Diagnostics
Deep Learning in Robotics
• Computer Vision, Speech Recognition and NLP
are direct applications in Robotics
• Training Robots to do specific tasks through
Deep Learning
– At UC Berkley: Train robot to perform tasks via
trial and error (e.x. screw a cap into water bottle)
Deep Learning in Robotics
• At Cornell: Deep Learning for detecting
Robotic Grasps (using Baxter)
Deep Learning for Detecting Robotic Grasps, Ian Lenz, Honglak Lee, Ashutosh Saxena. To
appear in International Journal of Robotics Research (IJRR), 2014.
http://pr.cs.cornell.edu/deepgrasping/
Challenges
• So far worked only in Patter-recognition
domains where there is good structural
patterns in the input data (Vision, Voice,
Language)
• With other kind of datasets (finance,
marketing, human behavior, biology), there
are not any known applications
Resources
Yann Lecun,
NYU, Facebook AI Research
Geoff Hinton,
Uni Torento, Google
Yoshua Bengio,
Uni Montreal
Andrew Ng,
Standford, Baidu Nando De Freitas,
Oxford, Deepmind
Key players for talks, lectures, papers, tutorials,
datasets, etc.
Thank you!

More Related Content

What's hot

TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...
TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...
TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...Simplilearn
 
Neural Networks
Neural NetworksNeural Networks
Neural NetworksAdri Jovin
 
Stuart russell and peter norvig artificial intelligence - a modern approach...
Stuart russell and peter norvig   artificial intelligence - a modern approach...Stuart russell and peter norvig   artificial intelligence - a modern approach...
Stuart russell and peter norvig artificial intelligence - a modern approach...Lê Anh Đạt
 
Introduction to Recurrent Neural Network
Introduction to Recurrent Neural NetworkIntroduction to Recurrent Neural Network
Introduction to Recurrent Neural NetworkYan Xu
 
Genetic Algorithm by Example
Genetic Algorithm by ExampleGenetic Algorithm by Example
Genetic Algorithm by ExampleNobal Niraula
 
What is the Expectation Maximization (EM) Algorithm?
What is the Expectation Maximization (EM) Algorithm?What is the Expectation Maximization (EM) Algorithm?
What is the Expectation Maximization (EM) Algorithm?Kazuki Yoshida
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer VisionSungjoon Choi
 
State space search and Problem Solving techniques
State space search and Problem Solving techniquesState space search and Problem Solving techniques
State space search and Problem Solving techniquesKirti Verma
 
Introduction to Artificial Intelligence
Introduction to Artificial IntelligenceIntroduction to Artificial Intelligence
Introduction to Artificial Intelligenceananth
 
Notes from Coursera Deep Learning courses by Andrew Ng
Notes from Coursera Deep Learning courses by Andrew NgNotes from Coursera Deep Learning courses by Andrew Ng
Notes from Coursera Deep Learning courses by Andrew NgdataHacker. rs
 
Machine Learning Foundations
Machine Learning FoundationsMachine Learning Foundations
Machine Learning FoundationsAlbert Y. C. Chen
 
Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual IntroductionLukas Masuch
 
Artificial Intelligence PEAS
Artificial Intelligence PEASArtificial Intelligence PEAS
Artificial Intelligence PEASMUDASIRABBAS9
 
Introduction For seq2seq(sequence to sequence) and RNN
Introduction For seq2seq(sequence to sequence) and RNNIntroduction For seq2seq(sequence to sequence) and RNN
Introduction For seq2seq(sequence to sequence) and RNNHye-min Ahn
 
Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Basit Rafiq
 
07 Machine Learning - Expectation Maximization
07 Machine Learning - Expectation Maximization07 Machine Learning - Expectation Maximization
07 Machine Learning - Expectation MaximizationAndres Mendez-Vazquez
 
Recurrent neural networks rnn
Recurrent neural networks   rnnRecurrent neural networks   rnn
Recurrent neural networks rnnKuppusamy P
 

What's hot (20)

TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...
TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...
TensorFlow Tutorial | Deep Learning With TensorFlow | TensorFlow Tutorial For...
 
Neural Networks
Neural NetworksNeural Networks
Neural Networks
 
Neural networks introduction
Neural networks introductionNeural networks introduction
Neural networks introduction
 
Stuart russell and peter norvig artificial intelligence - a modern approach...
Stuart russell and peter norvig   artificial intelligence - a modern approach...Stuart russell and peter norvig   artificial intelligence - a modern approach...
Stuart russell and peter norvig artificial intelligence - a modern approach...
 
Introduction to Recurrent Neural Network
Introduction to Recurrent Neural NetworkIntroduction to Recurrent Neural Network
Introduction to Recurrent Neural Network
 
Genetic Algorithm by Example
Genetic Algorithm by ExampleGenetic Algorithm by Example
Genetic Algorithm by Example
 
What is the Expectation Maximization (EM) Algorithm?
What is the Expectation Maximization (EM) Algorithm?What is the Expectation Maximization (EM) Algorithm?
What is the Expectation Maximization (EM) Algorithm?
 
Attention Models (D3L6 2017 UPC Deep Learning for Computer Vision)
Attention Models (D3L6 2017 UPC Deep Learning for Computer Vision)Attention Models (D3L6 2017 UPC Deep Learning for Computer Vision)
Attention Models (D3L6 2017 UPC Deep Learning for Computer Vision)
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer Vision
 
State space search and Problem Solving techniques
State space search and Problem Solving techniquesState space search and Problem Solving techniques
State space search and Problem Solving techniques
 
Introduction to Artificial Intelligence
Introduction to Artificial IntelligenceIntroduction to Artificial Intelligence
Introduction to Artificial Intelligence
 
Notes from Coursera Deep Learning courses by Andrew Ng
Notes from Coursera Deep Learning courses by Andrew NgNotes from Coursera Deep Learning courses by Andrew Ng
Notes from Coursera Deep Learning courses by Andrew Ng
 
Machine Learning Foundations
Machine Learning FoundationsMachine Learning Foundations
Machine Learning Foundations
 
Deep learning - A Visual Introduction
Deep learning - A Visual IntroductionDeep learning - A Visual Introduction
Deep learning - A Visual Introduction
 
Artificial Intelligence PEAS
Artificial Intelligence PEASArtificial Intelligence PEAS
Artificial Intelligence PEAS
 
Introduction For seq2seq(sequence to sequence) and RNN
Introduction For seq2seq(sequence to sequence) and RNNIntroduction For seq2seq(sequence to sequence) and RNN
Introduction For seq2seq(sequence to sequence) and RNN
 
Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)
 
Deep learning
Deep learningDeep learning
Deep learning
 
07 Machine Learning - Expectation Maximization
07 Machine Learning - Expectation Maximization07 Machine Learning - Expectation Maximization
07 Machine Learning - Expectation Maximization
 
Recurrent neural networks rnn
Recurrent neural networks   rnnRecurrent neural networks   rnn
Recurrent neural networks rnn
 

Viewers also liked

Network Intrusion Detection Dean Final, actual version
Network Intrusion Detection Dean Final, actual versionNetwork Intrusion Detection Dean Final, actual version
Network Intrusion Detection Dean Final, actual versionTianye Song
 
Impact of Artificial Intelligence/Machine Learning on Workforce Capability
Impact of Artificial Intelligence/Machine Learning on Workforce CapabilityImpact of Artificial Intelligence/Machine Learning on Workforce Capability
Impact of Artificial Intelligence/Machine Learning on Workforce CapabilityLearningCafe
 
Artificial intelligence in business
Artificial intelligence in businessArtificial intelligence in business
Artificial intelligence in businessNisha Choudhary
 
Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...
Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...
Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...Humanity Plus
 
Marek Rosa - Inventing General Artificial Intelligence: A Vision and Methodology
Marek Rosa - Inventing General Artificial Intelligence: A Vision and MethodologyMarek Rosa - Inventing General Artificial Intelligence: A Vision and Methodology
Marek Rosa - Inventing General Artificial Intelligence: A Vision and MethodologyMachine Learning Prague
 
Supermathematics and Artificial General Intelligence
Supermathematics and Artificial General IntelligenceSupermathematics and Artificial General Intelligence
Supermathematics and Artificial General IntelligenceJordan Bennett
 
Artificial general intelligence-libre
Artificial general intelligence-libreArtificial general intelligence-libre
Artificial general intelligence-libreSergio Viademonte
 
Whole Brain Connectomic Architecture to Develop General Artificial Intelligence
Whole Brain Connectomic Architecture to Develop General Artificial IntelligenceWhole Brain Connectomic Architecture to Develop General Artificial Intelligence
Whole Brain Connectomic Architecture to Develop General Artificial IntelligenceThe Whole Brain Architecture Initiative
 
From Narrow AI to Artificial General Intelligence (AGI)
From Narrow AI to Artificial General Intelligence (AGI)From Narrow AI to Artificial General Intelligence (AGI)
From Narrow AI to Artificial General Intelligence (AGI)Helgi Páll Helgason, PhD
 
Smart Data Webinar: Artificial General Intelligence - When Can I Get It?
Smart Data Webinar: Artificial General Intelligence - When Can I Get It?Smart Data Webinar: Artificial General Intelligence - When Can I Get It?
Smart Data Webinar: Artificial General Intelligence - When Can I Get It?DATAVERSITY
 
Artificial Intelligence in Business
Artificial Intelligence in BusinessArtificial Intelligence in Business
Artificial Intelligence in BusinessAjay Kumar
 
Augmented Reality, Artificial Intelligence, and Business Intelligence
Augmented Reality, Artificial Intelligence, and Business IntelligenceAugmented Reality, Artificial Intelligence, and Business Intelligence
Augmented Reality, Artificial Intelligence, and Business IntelligencePatrick
 
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017Carol Smith
 

Viewers also liked (16)

Network Intrusion Detection Dean Final, actual version
Network Intrusion Detection Dean Final, actual versionNetwork Intrusion Detection Dean Final, actual version
Network Intrusion Detection Dean Final, actual version
 
Deep Learning
Deep LearningDeep Learning
Deep Learning
 
Impact of Artificial Intelligence/Machine Learning on Workforce Capability
Impact of Artificial Intelligence/Machine Learning on Workforce CapabilityImpact of Artificial Intelligence/Machine Learning on Workforce Capability
Impact of Artificial Intelligence/Machine Learning on Workforce Capability
 
Artificial intelligence in business
Artificial intelligence in businessArtificial intelligence in business
Artificial intelligence in business
 
Artificial intelligence in business
Artificial intelligence in businessArtificial intelligence in business
Artificial intelligence in business
 
Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...
Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...
Is Deep-Layered Machine Learning the Catalyst for an Artificial General Intel...
 
Marek Rosa - Inventing General Artificial Intelligence: A Vision and Methodology
Marek Rosa - Inventing General Artificial Intelligence: A Vision and MethodologyMarek Rosa - Inventing General Artificial Intelligence: A Vision and Methodology
Marek Rosa - Inventing General Artificial Intelligence: A Vision and Methodology
 
Supermathematics and Artificial General Intelligence
Supermathematics and Artificial General IntelligenceSupermathematics and Artificial General Intelligence
Supermathematics and Artificial General Intelligence
 
Artificial general intelligence-libre
Artificial general intelligence-libreArtificial general intelligence-libre
Artificial general intelligence-libre
 
Whole Brain Connectomic Architecture to Develop General Artificial Intelligence
Whole Brain Connectomic Architecture to Develop General Artificial IntelligenceWhole Brain Connectomic Architecture to Develop General Artificial Intelligence
Whole Brain Connectomic Architecture to Develop General Artificial Intelligence
 
From Narrow AI to Artificial General Intelligence (AGI)
From Narrow AI to Artificial General Intelligence (AGI)From Narrow AI to Artificial General Intelligence (AGI)
From Narrow AI to Artificial General Intelligence (AGI)
 
Smart Data Webinar: Artificial General Intelligence - When Can I Get It?
Smart Data Webinar: Artificial General Intelligence - When Can I Get It?Smart Data Webinar: Artificial General Intelligence - When Can I Get It?
Smart Data Webinar: Artificial General Intelligence - When Can I Get It?
 
Artificial Intelligence in Business
Artificial Intelligence in BusinessArtificial Intelligence in Business
Artificial Intelligence in Business
 
Augmented Reality, Artificial Intelligence, and Business Intelligence
Augmented Reality, Artificial Intelligence, and Business IntelligenceAugmented Reality, Artificial Intelligence, and Business Intelligence
Augmented Reality, Artificial Intelligence, and Business Intelligence
 
The AI Rush
The AI RushThe AI Rush
The AI Rush
 
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017
AI and Machine Learning Demystified by Carol Smith at Midwest UX 2017
 

Similar to Deep Learning: Towards General Artificial Intelligence

Deep learning introduction
Deep learning introductionDeep learning introduction
Deep learning introductionAdwait Bhave
 
Deep learning short introduction
Deep learning short introductionDeep learning short introduction
Deep learning short introductionAdwait Bhave
 
Deep learning tutorial 9/2019
Deep learning tutorial 9/2019Deep learning tutorial 9/2019
Deep learning tutorial 9/2019Amr Rashed
 
Deep Learning Tutorial
Deep Learning TutorialDeep Learning Tutorial
Deep Learning TutorialAmr Rashed
 
Training machine learning deep learning 2017
Training machine learning deep learning 2017Training machine learning deep learning 2017
Training machine learning deep learning 2017Iwan Sofana
 
Yann le cun
Yann le cunYann le cun
Yann le cunYandex
 
Introduction to deep learning
Introduction to deep learningIntroduction to deep learning
Introduction to deep learningAmr Rashed
 
Deep learning with tensorflow
Deep learning with tensorflowDeep learning with tensorflow
Deep learning with tensorflowCharmi Chokshi
 
MLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learningMLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learningCharles Deledalle
 
Deep Learning For Practitioners, lecture 2: Selecting the right applications...
Deep Learning For Practitioners,  lecture 2: Selecting the right applications...Deep Learning For Practitioners,  lecture 2: Selecting the right applications...
Deep Learning For Practitioners, lecture 2: Selecting the right applications...ananth
 
Big Data Malaysia - A Primer on Deep Learning
Big Data Malaysia - A Primer on Deep LearningBig Data Malaysia - A Primer on Deep Learning
Big Data Malaysia - A Primer on Deep LearningPoo Kuan Hoong
 
An Introduction to Deep Learning
An Introduction to Deep LearningAn Introduction to Deep Learning
An Introduction to Deep LearningPoo Kuan Hoong
 
Deep learning: the future of recommendations
Deep learning: the future of recommendationsDeep learning: the future of recommendations
Deep learning: the future of recommendationsBalázs Hidasi
 
MLlecture1.ppt
MLlecture1.pptMLlecture1.ppt
MLlecture1.pptbutest
 
MLlecture1.ppt
MLlecture1.pptMLlecture1.ppt
MLlecture1.pptbutest
 

Similar to Deep Learning: Towards General Artificial Intelligence (20)

Deep learning introduction
Deep learning introductionDeep learning introduction
Deep learning introduction
 
Deep learning short introduction
Deep learning short introductionDeep learning short introduction
Deep learning short introduction
 
Deep learning tutorial 9/2019
Deep learning tutorial 9/2019Deep learning tutorial 9/2019
Deep learning tutorial 9/2019
 
Deep Learning Tutorial
Deep Learning TutorialDeep Learning Tutorial
Deep Learning Tutorial
 
Training machine learning deep learning 2017
Training machine learning deep learning 2017Training machine learning deep learning 2017
Training machine learning deep learning 2017
 
Yann le cun
Yann le cunYann le cun
Yann le cun
 
Introduction to deep learning
Introduction to deep learningIntroduction to deep learning
Introduction to deep learning
 
Deep learning with tensorflow
Deep learning with tensorflowDeep learning with tensorflow
Deep learning with tensorflow
 
Deep learning - a primer
Deep learning - a primerDeep learning - a primer
Deep learning - a primer
 
Deep learning - a primer
Deep learning - a primerDeep learning - a primer
Deep learning - a primer
 
MLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learningMLIP - Chapter 3 - Introduction to deep learning
MLIP - Chapter 3 - Introduction to deep learning
 
Deep Learning For Practitioners, lecture 2: Selecting the right applications...
Deep Learning For Practitioners,  lecture 2: Selecting the right applications...Deep Learning For Practitioners,  lecture 2: Selecting the right applications...
Deep Learning For Practitioners, lecture 2: Selecting the right applications...
 
tensorflow.pptx
tensorflow.pptxtensorflow.pptx
tensorflow.pptx
 
Big Data Malaysia - A Primer on Deep Learning
Big Data Malaysia - A Primer on Deep LearningBig Data Malaysia - A Primer on Deep Learning
Big Data Malaysia - A Primer on Deep Learning
 
An Introduction to Deep Learning
An Introduction to Deep LearningAn Introduction to Deep Learning
An Introduction to Deep Learning
 
Deep learning: the future of recommendations
Deep learning: the future of recommendationsDeep learning: the future of recommendations
Deep learning: the future of recommendations
 
Lecture 1 (bce-7)
Lecture   1 (bce-7)Lecture   1 (bce-7)
Lecture 1 (bce-7)
 
Deep learning.pptx
Deep learning.pptxDeep learning.pptx
Deep learning.pptx
 
MLlecture1.ppt
MLlecture1.pptMLlecture1.ppt
MLlecture1.ppt
 
MLlecture1.ppt
MLlecture1.pptMLlecture1.ppt
MLlecture1.ppt
 

Recently uploaded

BabyOno dropshipping via API with DroFx.pptx
BabyOno dropshipping via API with DroFx.pptxBabyOno dropshipping via API with DroFx.pptx
BabyOno dropshipping via API with DroFx.pptxolyaivanovalion
 
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girlCall Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girlkumarajju5765
 
Edukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFxEdukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFxolyaivanovalion
 
Capstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics ProgramCapstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics ProgramMoniSankarHazra
 
Accredited-Transport-Cooperatives-Jan-2021-Web.pdf
Accredited-Transport-Cooperatives-Jan-2021-Web.pdfAccredited-Transport-Cooperatives-Jan-2021-Web.pdf
Accredited-Transport-Cooperatives-Jan-2021-Web.pdfadriantubila
 
Zuja dropshipping via API with DroFx.pptx
Zuja dropshipping via API with DroFx.pptxZuja dropshipping via API with DroFx.pptx
Zuja dropshipping via API with DroFx.pptxolyaivanovalion
 
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...amitlee9823
 
Data-Analysis for Chicago Crime Data 2023
Data-Analysis for Chicago Crime Data  2023Data-Analysis for Chicago Crime Data  2023
Data-Analysis for Chicago Crime Data 2023ymrp368
 
Halmar dropshipping via API with DroFx
Halmar  dropshipping  via API with DroFxHalmar  dropshipping  via API with DroFx
Halmar dropshipping via API with DroFxolyaivanovalion
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusTimothy Spann
 
Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxolyaivanovalion
 
Carero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxCarero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxolyaivanovalion
 
Cheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 night
Cheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 nightCheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 night
Cheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 nightDelhi Call girls
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxfirstjob4
 
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130Suhani Kapoor
 
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779Delhi Call girls
 

Recently uploaded (20)

BabyOno dropshipping via API with DroFx.pptx
BabyOno dropshipping via API with DroFx.pptxBabyOno dropshipping via API with DroFx.pptx
BabyOno dropshipping via API with DroFx.pptx
 
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girlCall Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
Call Girls 🫤 Dwarka ➡️ 9711199171 ➡️ Delhi 🫦 Two shot with one girl
 
Edukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFxEdukaciniai dropshipping via API with DroFx
Edukaciniai dropshipping via API with DroFx
 
Capstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics ProgramCapstone Project on IBM Data Analytics Program
Capstone Project on IBM Data Analytics Program
 
Accredited-Transport-Cooperatives-Jan-2021-Web.pdf
Accredited-Transport-Cooperatives-Jan-2021-Web.pdfAccredited-Transport-Cooperatives-Jan-2021-Web.pdf
Accredited-Transport-Cooperatives-Jan-2021-Web.pdf
 
Zuja dropshipping via API with DroFx.pptx
Zuja dropshipping via API with DroFx.pptxZuja dropshipping via API with DroFx.pptx
Zuja dropshipping via API with DroFx.pptx
 
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
Junnasandra Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalore...
 
Sampling (random) method and Non random.ppt
Sampling (random) method and Non random.pptSampling (random) method and Non random.ppt
Sampling (random) method and Non random.ppt
 
Data-Analysis for Chicago Crime Data 2023
Data-Analysis for Chicago Crime Data  2023Data-Analysis for Chicago Crime Data  2023
Data-Analysis for Chicago Crime Data 2023
 
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
 
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in  KishangarhDelhi 99530 vip 56974 Genuine Escort Service Call Girls in  Kishangarh
Delhi 99530 vip 56974 Genuine Escort Service Call Girls in Kishangarh
 
Halmar dropshipping via API with DroFx
Halmar  dropshipping  via API with DroFxHalmar  dropshipping  via API with DroFx
Halmar dropshipping via API with DroFx
 
Generative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and MilvusGenerative AI on Enterprise Cloud with NiFi and Milvus
Generative AI on Enterprise Cloud with NiFi and Milvus
 
Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptx
 
Carero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxCarero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptx
 
Cheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 night
Cheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 nightCheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 night
Cheap Rate Call girls Sarita Vihar Delhi 9205541914 shot 1500 night
 
CHEAP Call Girls in Saket (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Saket (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICECHEAP Call Girls in Saket (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Saket (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptx
 
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
 
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
Best VIP Call Girls Noida Sector 39 Call Me: 8448380779
 

Deep Learning: Towards General Artificial Intelligence

  • 1. Deep Learning: Towards General Artificial Intelligence Dr. Rukshan Batuwita (Machine Learning Scientist) Senior Data Scientist Ambiata Pvt Ltd, Sydney, Australia
  • 2. What is Artificial Intelligence? • Field of Study to Develop Machines that act like Humans! • Recent Definition: Develop Machines that act rationally!
  • 3. General AI • Learning and Reasoning • Planning • Adaptability • Vision • Speech recognition • Automation • Mobility • etc.
  • 4. Narrow AI • Learning and Reasoning • Planning • Adaptability • Vision • Speech recognition • Automation • Mobility • etc. • People have been working on for the last 50 years • This has many applications Narrow AI
  • 5. What is Machine Learning? • Computer Program: Input to output mapping Computer Program (Algorithm/List of Instructions) Inputs Outputs When we know the algorithm to solve a task, then we can program it ?Inputs Outputs AI Problems: Ex. ? Cat or Dog
  • 6. In Machine Learning… Algorithm (Model) Cat or Dog Machine Learning techniques . . . . . . Train/Learn
  • 8. Biological Neurons Inputs Outputs Biological Brain Biological Neural Network Biological Neuron Processing/Computing
  • 9. Biological Learning • Biological Neuron Learning happens due to some chemical reactions in synaptic connections A synaptic connection A typical adult human brain has about 1014 synapses (connections)
  • 10. Artificial Neuron • A computational model Y y = f ( wj xj j=1 d å ) • Called ‘Perceptron’ • Introduced in 1960’s • Weights can be learned by an optimization method like Gradient Descent Weights (represent chemicals) Inputs Outputs Processing /Computing Inputs Inputs
  • 11. Perceptron Perceptron = Linear Regression Perceptron = Logistic Regression Activation Functions: Y y = f ( wj xj j=1 d å )
  • 12. Artificial Neural Network weights weights • Artificial Neurons are corrected together to form a network • Called Multi Layer Perceptron (MLP) • A Non-linear model of the parameters • Trained by popular Backpropagation (Gradient Descent)
  • 13. Backpropagation – Main Idea 1. Calculate Error/Loss = f(Label , Prediction) 2. Calculate Gradient/Derivative of the Loss w.r.t. each weight 3. In order to calculate the gradient of the inner weights, apply the chain rule of derivatives 4. Update each weight in the direction of the negative gradient (Gradient Descent) Error = f(label, prediction)
  • 15. • Quite popular in 1980’s and 1990’s • Worked well for some pattern recognition problems: – Ex: Handwritten digit recognition Le-Net used by US postal department • Other ML methods (ex. Kernel methods such as SVMs) dominated ANNs in early 2000’s • Main problems of ANNs: – Local-minima (since the loss function is non-convex) – Difficult to train networks with more then 3/4 layers • Overfitting • Computational time • Vanishing Gradient problem (e.g. when Sigmoid activation is used) • (didn’t work well in more complex problems like general image classification) Before 2006… (LeCun et al., 1998) Yan LeCun, NYU Geoff Hinton, Uni Torento Yoshua Bengio, Uni Montreal
  • 16. After 2006… • Several major breakthroughs happened giving birth to Deep Learning • In general, Deep Learning is nothing but good old Neural Networks with many layers: … N … • Deep Learning methods have been significantly outperforming the existing methods in major Computer Vision and Speech Recognition competitions since 2010
  • 17. ImageNet Results… About 14M images of 100k categories/concepts
  • 18. Main Advancements made Deep Learning possible
  • 19. 1. Unsupervised Feature Learning • In classical Machine Learning: Feature ExtractionRaw Data Feature pre- processing Model Learning 80%-90% of the effort (Human effort) • In Deep Learning: Feature LearningRaw Data Model Learning Deep Learning + Model Model Feature Learning = Representation Learning = Embedding Learning
  • 20. Feature Learning/Representation Learning (Ex. Face Detection) Layer 1 (Detects Edges) Layer 2 (Detects Face parts Combination of edges) Deeper layer (Detects Faces) Input Pixels InputPixels
  • 21. Techniques for Representation Learning 1. Layer-wise unsupervised pre-training 1. Stacked Autoencoders Input Output Encode Decode Edge Detectors Autoencoder • No labels required • Unsupervised Training Pixelinput Pixeloutput
  • 22. Stacked Autoencoders 1. Train one layer autoencoder at a time [unsupervised learning] and stack them 2. Then train the final network using the available labels [supervised learning] Low level features Higher level features Higher level features INPUT LABEL Techniques for Representation Learning Input
  • 23. 1. Layer-wise unsupervised pre-training 2. Deep Belief Networks (Restricted Boltzmann MMachines (RBM) are stacked together) Techniques for Representation Learning
  • 24. Techniques for Representation Learning 2. Deep Convolution Networks Convolution Filters Kernel/convolution matrix/mask/filter Edge Detector X_1 … … … … … … … X_9 W_1 … … … … … … … W_9 zi = xiwi i=1 9 å X 3x3 Image patch Z CONV( ),
  • 25. Techniques for feature learning 2. Deep Convolution Networks Feature Extraction Classification • Convolutional Filters (low-level and high-level) are also learned automatically with Backprop Subsampling = average, max (max pooling) - noise reduction Different types of filters result in different feature maps
  • 26. Techniques for feature learning 2. Deep Convolution Networks Inputlayer W_1 x1 x1 x2 x3 x4 W_2 W_3 W_4 x5 x6 x2 x3 x4 x5 x6 W_1 W_2 W_3 W_4 … … … … … … W_1 W_2 W_3 W_4 2X2 filter x5 x6 … … … … … … W_1 W_2 W_3 W_4 x1 x3 … … … … … … W_1 W_2 W_3 W_4 • Each layer is represented by connected neurons • Each convolution layer is connected to the previous layers sparsely and with shared weights
  • 27. Techniques for feature learning 2. Deep Convolution Networks • Convolution and Subsampling (Pooling) leads to detect translational invariance features • Works with language (document classification, translation) and Voice recognition
  • 29. Motivations for Feature/Representation learning 1. Cut down the effort of handcrafting features 2. Hierarchical, distributed, compositional knowledge representations in Brain – Humans organize their concepts and ideas hierarchically – Humans first learn simple concepts and compose them together to represent complex ideas – Human problem solving/Engineering (multiple level of abstractions) – Human language understanding – Pattern recognition in brain, etc.
  • 30. Motivations for Feature/Representation learning • Hierarchical, distributed, compositional knowledge representation/pattern recognition in Brain Pattern Recognition in Brain Pattern Recognition In Deep Learning
  • 31. Motivations for Feature/Representation learning 3. Power of distributed, compositional representations • Concepts are represented as composition of features at different levels • The number of concepts can be represented grow exponentially with the size of the network Input Low-level representations (e.x. edges) High-level representations
  • 32. Motivations for Feature/Representation learning 4. Manifold Learning • Assumption: Input data has some structure (not 100% random) which is concentrated in a lower-dimensional manifold of the original features • Ex: most of the arbitrary pixel value configurations don’t create the images of faces • Representation in each layer can be considered as a learned manifold of the previous layer 28! F or AI T a sk s: Ma n i f ol d st r uct ur e • examples!concentrate!near!a!lower!dimensional!“manifold! • Evidence:$most$input$configuraDons$are$unlikely$ Pixels (32*32 image) E.x.
  • 33. Motivations for Feature/Representation learning 5. Transfer Learning – Generalization: ability of a model to predict well on unseen test data – Representation of complex concepts -> Deep Networks – Good generalization of complex models like Deep Neural Networks rely on the availability of large number of labeled training data – Most of the available data are not labeled – In Transfer Learning 1. Train a Deep Network with unlabeled data in unsupervised manner 2. Use the available labeled data to train the required model
  • 34. Motivations for Feature/Representation learning 5. Transfer Learning Example: Image recognition model . . . Unsupervised pre-training with unlabeled data to learn the representations of different levels of abstraction Transfer the knowledge car Supervised Learning with available labeled data ... Hu man
  • 35. Variations of Transfer Learning • Multi-Instance Learning (when labels are not available at the instance level) Document Classification Model Based on the similarity of the sentence/word embedding [Kotzias, Denil and deFreitas, 2014]
  • 36. Variations of Transfer Learning • Max-margin Learning without labels [From machine learning to machine reasoning, Leon Bottou, 2014]
  • 37. Variations of Transfer Learning • Max-margin Learning without labels [NLP almost from scratch, Ronan Collobert et al., 2011]
  • 38. Other advancements made Deep Learning possible
  • 39. Other advancements… • ‘Dropouts’ regularization for training with Backpropagation for higher generalization • Rectified Linear Functions instead of Sigmoid (avoid vanishing gradient problem)
  • 40. Other architectures… • Memory Networks (LSTM) – Question answering • Recurrent Networks – Detecting inputs with sequential relationships (voice recognition) • Combination of existing architectures
  • 41. Improved Computing Power… GPU Computation – Parallel Neural Network Training on GPU clusters (ideal for simple Matrix/Vector operations, hence for backpropagation) – Reduced the training time of deep networks from weeks to days – NVIDIA CUDA Deep Neural Network library
  • 42. Improved Computing Power… • Commodity Hardware – Multi-core single machines, clusters, GUP clusters • Open source software – Torch (open source ML library, https://github.com/torch/torch7/) – From Yoshua Bengio’s group http://deeplearning.net/software/theano/) – Caffe – Google TensorFlow
  • 43. Industrial Applications of Deep Learning Techniques
  • 44. Google Brain Project – Started by Andrew Ng in 2011 – In 2012: Neural Network with 1 Billion connection was trained across 16,000 CPU cores – They considered this ANN as simulating a very small- scale “newborn brain,” and show it YouTube video for a week, what will it learn? – Used an Unsupervised (Self-taught-learning) to learn features from unlabeled Google images – Autoencoder – Exposed to fames of 10M YouTube videos over a week http://googleblog.blogspot.com.au/2012/06/using-large-scale-brain-simulations-for.html http://static.googleusercontent.com/media/research.google.com/en//archive/unsupervised_icml2012.pdf Andrew Ng, Standford
  • 45. Google Brain Project What Happened? • One of the artificial neurons learned to respond strongly to pictures of Cats.
  • 46. Evolution of Deep Leaning at Google – Google has been heavily investing on Deep Learning research – In 2013 Google hired Geoff Hinton and acquired his start-up company DNNResearch Inc. – In 2014 they purchased a UK-based Machine Learning company called DeepMind Technologies for estimated $650 Million
  • 47. Deep Mind Apollo Program for AI Working towards solving General AI with Deep Reinforcement Learning….
  • 48. DeepMind • Famous paper: Applying Deep RL to train agents to play classic Atari games
  • 50. AlphaGo • Traditional Chinese game - Go • The most complex board game of all • Alpha Go beat the world champion in Go 4/5 Lee Sedol
  • 51. Deep Dream (http://deepdreamgenerator.com) • What features will be picked up by Google’s Deep ANNs? Deep ANN Original Image Original Image + Recognized Features
  • 52. Google Voice Recognition (in Android and Search by Voice) Deep Learning Products at Google Google search by Image (Search for similar images to an uploaded image)
  • 53. Facebook • Yann Lecun is the head of Facebook AI Research • Face Recognition: Deep Face • claim to have close to human-level performance • Personal Assistant: Facebook M
  • 54. Other… • Microsoft Cortana, Skype Translate • Nvida Self Driving Cars • Image Captioning Systems • Siemens Medical Image Diagnostics
  • 55. Deep Learning in Robotics • Computer Vision, Speech Recognition and NLP are direct applications in Robotics • Training Robots to do specific tasks through Deep Learning – At UC Berkley: Train robot to perform tasks via trial and error (e.x. screw a cap into water bottle)
  • 56. Deep Learning in Robotics • At Cornell: Deep Learning for detecting Robotic Grasps (using Baxter) Deep Learning for Detecting Robotic Grasps, Ian Lenz, Honglak Lee, Ashutosh Saxena. To appear in International Journal of Robotics Research (IJRR), 2014. http://pr.cs.cornell.edu/deepgrasping/
  • 57. Challenges • So far worked only in Patter-recognition domains where there is good structural patterns in the input data (Vision, Voice, Language) • With other kind of datasets (finance, marketing, human behavior, biology), there are not any known applications
  • 58. Resources Yann Lecun, NYU, Facebook AI Research Geoff Hinton, Uni Torento, Google Yoshua Bengio, Uni Montreal Andrew Ng, Standford, Baidu Nando De Freitas, Oxford, Deepmind Key players for talks, lectures, papers, tutorials, datasets, etc.