SlideShare a Scribd company logo
1 of 58
Download to read offline
Copyright © SAS Institute Inc. All rights reserved.
Deep Learning and What’s Next?
Tao Wang (t.wang@sas.com)
2018
AI-Now
Copyright © SAS Institute Inc. All rights reserved.
Part 1:
Introduction
Copyright © SAS Institute Inc. All rights reserved.
AI vs. Machine Learning
Take 1
3
Source: https://towardsdatascience.com/cousins-of-artificial-intelligence-dda4edc27b55
Copyright © SAS Institute Inc. All rights reserved.
AI vs. Machine Learning
Take 2
Machine learning is a “field of study that gives computers the
ability to learn without being explicitly programmed.”
– Arthur Samuel, 1959
4
“Artificial intelligence (AI), sometimes called machine
intelligence, is intelligence demonstrated by machines.”
– Wikipedia, retrieved 2018
Copyright © SAS Institute Inc. All rights reserved.
AI vs. Machine Learning
5
Take 3
Copyright © SAS Institute Inc. All rights reserved.
AI vs. Machine Learning
Take 4 – my own version
6
AI: goal
Analytics: business Machine Learning: means
Copyright © SAS Institute Inc. All rights reserved.
Styles of Machine Learning
Other styles from different perspectives (active learning, transfer learning,
multi-task learning, adversarial learning, …)
Supervised Semi-SupervisedUnsupervised Reinforcement
• Training data
has labelled
target
• Predict label
for unseen
data
• Labels are known
for a subset of data
• A blend of
supervised and
unsupervised
learning
• Labels unknown
• Find patterns
and gain
insights from
the data
• An agent selects
actions to
maximize reward
in an
environment
• Face detection,
fraud
detection,
patient
identification
• Customer
clustering
• Association rule
mining
• Pre-processing for
supervised learning
to reduce labelling
cost and enhance
accuracy
• Game AI
• Robotics
7
Source: [5] X. Hunt, et al.
Copyright © SAS Institute Inc. All rights reserved.
What Can Machine Learning Do?
And so many other things!
Prediction
Decision and
Policy-Making
Data
Exploration
Rule
Learning
• Classification
• Regression
• Clustering
• Dimension reduction
• Anomaly detection
• Feature engineering
• Identifying
relational rules
within data
• Association rule
mining
• Supervised
learning
• Semi-supervised
learning
• Often unsupervised
learning
• Also supervised and
semi-supervised
learning
• Reinforcement
learning
• Supervised
learning
• Unsupervised
learning
• Semi-supervised
learning
• Learning through
trial and error to
identify best action
• Game playing
• Control problems
8
Source: [5] X. Hunt, et al.
Copyright © SAS Institute Inc. All rights reserved.
9
Machine Learning and Deep Learning
Machine
Learning
Deep
Learning
AlphaGo
Bidirectional Encode
AlphaFold
Image sources: Siliconangle, googleblog, profacgen
Copyright © SAS Institute Inc. All rights reserved.
Part 2:
Deep Learning = Deep
Neural Networks
(DL = DNN)?
10
Copyright © SAS Institute Inc. All rights reserved.
11
Deep Learning: Model with Depth
Shallow
Deep
Learning
• Model with one or a
few layers
• Multiple layers, layer-by-layer
processing
• Feature extraction/transformation
• Learn complex structures
Data
Model
Output
Data
Output
Model
Layer
Layer
Layer
Deep Learning (DL) = Deep Neural Networks (DNN), ignoring subtle stuff
Source: [5] X. Hunt, et al.
Copyright © SAS Institute Inc. All rights reserved.
Pros and cons
• Advantages
1. Requires minimal feature engineering (end-to-end ML)
2. Flexible structures
3. Learning often improves with more data
4. Proven track records in speech/text processing and image/video recognition
• Disadvantages
1. Difficult to interpret – often treated as a “black-box” model
2. Long training time, over-fitting
3. Hard to train, non-repeatable results, numerous architectures/hyper-parameters
4. Requires a large amount of training data to get good models
12
Copyright © SAS Institute Inc. All rights reserved.
Why so popular?
1. End2end/distributed feature learning
2. Advances in algorithms/optimizations (min-batch, drop-out, BN, SGD, etc.)
3. Cloud computing and GPU made it possible to train very deep models
4. Proven track records in speech/text processing and image/video recognition
13
Source: [6] D. Silver
Copyright © SAS Institute Inc. All rights reserved.
More about DNN
• When should I use DNN?
• Deal with image/video/text/speech
• Works for small-medium data, but prefers big data
• The underlying model is complex and non-linear
• OK with non-interpretability, and/or have cloud/GPU
• Common DNN architectures
• Deep Forward Nets
• Convolutional neural networks (CNN)
• Recurrent neural networks (RNN)
• Stacked auto-encoders
14
Copyright © SAS Institute Inc. All rights reserved.
Deep Forward Net
• A flat architecture
• Regression and classification
DNN
architectures
1
15
Source: [4] W. Thompson
Copyright © SAS Institute Inc. All rights reserved.
Convolutional neural network (CNN)
• A feedforward neural net with conv layers
• 3D volumes of neurons
• Feature extraction
• Applications: image/video recognition, NLP
DNN
architectures
2
16
Source: [4] W. Thompson
Copyright © SAS Institute Inc. All rights reserved.
Recurrent neural network (RNN)
• Contain at least one feed-back connection
• Time-series forecasting, speech recognition
DNN
architectures
3
delay
h1(t)h1(t-1)
17
Source: [4] W. Thompson
Copyright © SAS Institute Inc. All rights reserved.
Auto-encoder
• A generative graphical model
• Feature coding, dimension reduction and compression
DNN
architectures
4
18
Source: [4] W. Thompson
Copyright © SAS Institute Inc. All rights reserved.
DNN supported by SAS
19
Source: [7] White paper: How to Do Deep Learning With SAS?
Copyright © SAS Institute Inc. All rights reserved.
SAS platform for DL
20
Copyright © SAS Institute Inc. All rights reserved.
SAS® Visual Machine learning and Machine Learning
(VDMML)
Visual “drag & drop” GUI
21
Copyright © SAS Institute Inc. All rights reserved.
Applications of SAS Deep Learning
22
Source: [7] White paper: How to Do Deep Learning With SAS?
Copyright © SAS Institute Inc. All rights reserved.
Applications
Input
DNN
Military
Surveillance
Speech
recognition
Fraud
Detection
Image
classification
Autonomous
Vehicles
Patient
Identification
23
Source: [4] W. Thompson
Copyright © SAS Institute Inc. All rights reserved.
Autonomous vehicles
An application of DNN
The tipping point: level 3 Partial Autonomy
Source: https://iq.intel.com/autonomous-cars-road-ahead/
Expected Timeline for Full Autonomy?
Source: https://thelastdriverlicenseholder.com/2016/12/29/expected-timeline-for-full-autonomy/
Focus on Level 3 and deliver!
24
Copyright © SAS Institute Inc. All rights reserved.
Navigant Research Leaderboard
Automated Driving Vehicles
Source: https://www.navigantresearch.com/research/navigant-research-leaderboard-automated-driving-vehicles
25
Copyright © SAS Institute Inc. All rights reserved.
End to End Learning for Self-Driving Cars
Finally, you are talking about a paper…
• arXiv:1604.07316, Apr 2016, from NVIDIA
• Basic idea: behavioral cloning, train the car to drive like you do
• Uses CNN to map images from cameras to steering commands
• Never explicitly train the CNN to detect/follow lanes, path planning, etc.
26
High-level view of the data collection system Training the CNN Self-driving
Source: [1] M. Bojarski, et al.
Copyright © SAS Institute Inc. All rights reserved.
CNN architecture and the core source code
27
Read it from bottomup. Input layer, normalization layer, 5 conv2D layers: feature extraction. 3
fully-connected layers, output: controller.
27M connections, 250K parameters, 3MB in size. Source: arXiv:1604.07316
Source: github, the NVIDIA 2016 paper implementation
Copyright © SAS Institute Inc. All rights reserved.
Part 3:
What’s next?
28
THE POWER OF
THE PACK
AI with
THE POWER OF
DIVERSITY
AI with
THE POWER OF
TRUST
AI with
Copyright © SAS Institute Inc. All rights reserved.
29
Rediscover Deep Learning
End to
End
1
Distributed
Feature
Learning
2
Big Data
Big Model
3
Copyright © SAS Institute Inc. All rights reserved.
30
Source: Yoshua Bengio
Source: Pablo Picasso
Capsule Networks – power of the pack
Source: CB Insights, State of AI Source: Forbes
Copyright © SAS Institute Inc. All rights reserved.
Capsule Network paper
Yep, talking about paper again
• S. Sabour, N. Frosst, G. Hinton, Dynamic Routing Between Capsules,
Google Brain, NIPS 2017, https://arxiv.org/abs/1710.09829
• Introduced years ago by Hinton, but was not working properly until now
• Widely considered as the beginning of a new chapter of deep learning
• Some follow-up papers, such as Matrix Capsules With EM Routing
• https://openreview.net/pdf?id=HJWLfGWRb, ICLR 2018
• Introduced capsule convolution layer and more sophisticated routing
31
Source: http://www.cs.toronto.edu/~hinton
Copyright © SAS Institute Inc. All rights reserved.
Dynamic Routing Between Capsules
• Idea #1: capsule is an encapsulated vector/matrix in the network
• A capsule is a group of neurons that represents the parameters of some specific feature.
• A vector or matrix is extended from a scalar
• The length represents the probability of the presence of a feature or an object
• Each dimension within the capsule represents the detailed information of location, size,
orientation, etc.
• Idea #2: routing by agreement
• Lower-level capsule (which is near input) prefers to send its output to higher-level (which
is near output) capsules with “similar” prediction
• Cosine similarity is used to measure the agreement
32
Copyright © SAS Institute Inc. All rights reserved.
CapsNet Architecture
▪ Input: MNIST dataset
▪ ReLU conv1: extracts local features
▪ PrimaryCaps: forms new neural unit (capsule)
▪ DigitCaps: contains 10 capsules (number 0 to 9)
▪ Cosine similarity (routing) is applied between PrimaryCaps and DigitCaps
▪ Reconstruction: a regularization method to encourage the capsules to encode the input digit
Figure 1: A simple CapsNet with 3 layers Figure 2: Reconstruct a digit from the DigitCaps layer representation
source: https://arxiv.org/abs/1710.09829
33
Copyright © SAS Institute Inc. All rights reserved.
Core source code
Source: github, the NIPS 2017 paper implementation
34
Copyright © SAS Institute Inc. All rights reserved.
Numerical results of the NIPS paper
source: https://arxiv.org/abs/1710.09829
35
Copyright © SAS Institute Inc. All rights reserved.
36
𝐸 = 𝐸 − 𝐷
Deep Forest – power of diversity
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper series
• Deep Forest [10], using RF to do DL with the “3 key ingredients”:
• In-model feature extraction and transformation, end-to-end machine learning
• Layer by layer processing, distributed representation learning
• Complex model
• AutoEncoder by Forest [11]
• The first tree ensemble based auto-encoder
• Multi-Layered Gradient Boosting Decision Trees [12]
• A variant of target propagation, pseudo-mapping F, pseudo-inverse-mapping G,
pseudo-label Z (F-G-Z framework)
• More to come?
37
Why always Neural Nets? We can do DL using Decision Trees!
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Can you stop talking about papers? No, papers are fun, ☺
• IJCAI 2017 paper [10], by Zhou and his student
• DeepForest = Forest ensemble, double-happiness (ensemble of
ensembles)
1. Multi-grain scanning, sliding window to extract features
2. Cascade of multiple random forests layers, for prediction
• Very few hyper-parameters (how nice!) & as good as DNN
• Default settings are good for many applications
• Non-differentiable model, no back propagation
38
Source: https://en.wikipedia.org/wiki/Zhi-Hua_Zhou
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Problems of DNN
• Too many hyper-parameters (like an art rather than science)
• Does not work well for small data
• Model architecture/complexity is determined in advance (via tuning)
• Often overly complicated
• Shortcut connection, pruning, binarization, etc. are often applied
39
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Why deep forest? Motivations?
• Decision trees
• Architecture learning (grow/split until done)
• Data driven
• Almost unbeatable on tabular data in Kaggle
• Motivations
• DL = DNN?
• Can we do DL with non-differentiable models (no back-propagation)?
• Maybe repeatable results (think SGD)?
40
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Inspiration from DNN
• Distributed representation learning (end to end, in-model feature trans.)
• Layer-by-layer processing
• Model complexity
41
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Multi-Grained Scanning for Feature Engineering
42
• Sequential
relationships
are
important
• Spatial
relationships
are
important
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Cascade Forest Structure for Prediction
• Ensemble of
ensembles
• K-fold cross
validation
• Architecture
learning (stop
growing
when
satisfied)
43
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Class Vector Generation
44
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Overall Architecture
45
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Hyper-parameters and default settings
46
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Experimental results
47
Image Categorization Face Recognition
Music Classification Hand Movement Recognition
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
More experimental results
48
Sentiment Classification
Low-Dimensional Data
High-Dimensional Data
(hard to beat successful method
at its killer-app with
a brand-new algorithm)
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Running time
• PC with 2 Intel E5 2695 v4 CPUs (18 cores)
• IMDB dataset (25,000 examples, with 5,000 features)
• Deep Forest: 40 minutes
• DNN: can take over 60 minutes
49
Copyright © SAS Institute Inc. All rights reserved.
Deep Forest paper
Hyper-parameter sensitivity
50
Source: [10] Z. Zhou and J. Feng
Copyright © SAS Institute Inc. All rights reserved.
51
AI
Analytics Machine Learning
Blockchain – power of trust
Source: pixabay
Copyright © SAS Institute Inc. All rights reserved.
A Unified Analytical Framework for Trustable Machine Learning and
Automation Running with Blockchain
52
Source: [14] T. Wang
Copyright © SAS Institute Inc. All rights reserved.
Further reading list
• Reinforcement learning (play, explore, control, interact)
• An agent selects actions to maximize reward in an environment
• AI = Deep RL (D. Silver, 2016) vs. RL does not really work (I. Goodfellow, 2018)
• Generative adversarial networks (GAN) [9]
• Unsupervised learning using supervised learning as sampling model
• Infers models in a competing game with Generator (G) and Discriminator (D)
• Provides an attractive alternative to maximum likelihood techniques.
• Y. LeCun: “…There are many interesting development in deep learning…The most important one, …, is adversarial training….”
• Adaptive Neural Trees (ANT), https://arxiv.org/abs/1807.06699, 2018
- NN: end2end/distributed representation learning with pre-specified architecture, image/sequence
- DT: architecture learning with pre-specified features, tabular data
• BERT – Bidirectional Encoding model
• AlphaFold
53
Copyright © SAS Institute Inc. All rights reserved.
AI winter is coming?
54
Source: https://blog.piekniewski.info/2018/05/28/ai-winter-is-well-on-its-way/
Source: Google trend
Copyright © SAS Institute Inc. All rights reserved.
Closing Remarks
AI and machine learning are very hard – just keep trying!
55
Copyright © SAS Institute Inc. All rights reserved.
Selected References
• [1] M. Bojarski, et al., End to End Learning for Self-Driving Cars, arXiv:1604.07316, 2016.
• [2] S. Sabour, N. Frosst, G. Hinton, Dynamic Routing Between Capsules, Google Brain, NIPS 2017, https://arxiv.org/abs/1710.09829
• [3] D. Silver, A. Huang, et, al. (2016). "Mastering the game of Go with deep neural networks and tree search". Nature 529 (7587): 484–
489.
• [4] W. Thompson, et al., Introduction to Deep learning, SAS, 2016.
• [5] X. Hunt, et al., Machine Learning Landscape, SAS, 2017.
• [6] D. Silver, Tutorial: Deep Reinforcement Learning, 2017.
• [7] White paper: How to Do Deep Learning With SAS? 2018.
• [8] Y. LeCun, et al., Deep learning, Nature, 2015.
• [9] I. Goodfellow, et al., Generative Adversarial Net, https://arxiv.org/abs/1406.2661
• [10] Z. Zhou and J. Feng, Deep Forest, IJCAI 2017.
• [11] J. Feng and Z. Zhou, AutoEncoder by Forest, AAAI 2018.
• [12] J. Feng, Y. Yu, Z. Zhou, Multi-Layered Gradient Boosting Decision Trees, https://arxiv.org/abs/1806.00007, 2018
• [13] R. Tanno, et al., Adaptive Neural Trees, https://arxiv.org/abs/1807.06699, 17 Jul 2018.
• [14] T. Wang, A Unified Analytical Framework for Trustable Machine Learning and Automation Running with Blockchain, IEEE Big Data
Workshops, 2018.
56
Copyright © SAS Institute Inc. All rights reserved.
Running for 2019 ACM SIGAI Vice-Chair
Shameless election/campaign ads
Vote for Tao Wang
57
Copyright © SAS Institute Inc. All rights reserved.
Guest Speaker
Dr. Kirk Borne
58
Source: https://twitter.com/KirkDBorne

More Related Content

What's hot

KERAS Python Tutorial
KERAS Python TutorialKERAS Python Tutorial
KERAS Python TutorialMahmutKAMALAK
 
Introduction to Keras
Introduction to KerasIntroduction to Keras
Introduction to KerasJohn Ramey
 
Deep Learning Made Easy with Deep Features
Deep Learning Made Easy with Deep FeaturesDeep Learning Made Easy with Deep Features
Deep Learning Made Easy with Deep FeaturesTuri, Inc.
 
AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...Apache MXNet
 
Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015
Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015
Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015Turi, Inc.
 
Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017
Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017
Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017StampedeCon
 
Artificial Intelligence, Machine Learning and Deep Learning
Artificial Intelligence, Machine Learning and Deep LearningArtificial Intelligence, Machine Learning and Deep Learning
Artificial Intelligence, Machine Learning and Deep LearningSujit Pal
 
Deep Learning: Chapter 11 Practical Methodology
Deep Learning: Chapter 11 Practical MethodologyDeep Learning: Chapter 11 Practical Methodology
Deep Learning: Chapter 11 Practical MethodologyJason Tsai
 
AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...Vandana Kannan
 
Introduction to deep learning
Introduction to deep learningIntroduction to deep learning
Introduction to deep learningAmr Rashed
 
Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...
Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...
Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...Greg Makowski
 
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 Separating Hype from Reality in Deep Learning with Sameer Farooqui Separating Hype from Reality in Deep Learning with Sameer Farooqui
Separating Hype from Reality in Deep Learning with Sameer FarooquiDatabricks
 
Intro to Mahout -- DC Hadoop
Intro to Mahout -- DC HadoopIntro to Mahout -- DC Hadoop
Intro to Mahout -- DC HadoopGrant Ingersoll
 
Fundamental of deep learning
Fundamental of deep learningFundamental of deep learning
Fundamental of deep learningStanley Wang
 
Introduction to machine learning
Introduction to machine learningIntroduction to machine learning
Introduction to machine learningSanghamitra Deb
 
Artificial Intelligence and Deep Learning in Azure, CNTK and Tensorflow
Artificial Intelligence and Deep Learning in Azure, CNTK and TensorflowArtificial Intelligence and Deep Learning in Azure, CNTK and Tensorflow
Artificial Intelligence and Deep Learning in Azure, CNTK and TensorflowJen Stirrup
 
Convolutional Neural Networks at scale in Spark MLlib
Convolutional Neural Networks at scale in Spark MLlibConvolutional Neural Networks at scale in Spark MLlib
Convolutional Neural Networks at scale in Spark MLlibDataWorks Summit
 
Transfer Learning: An overview
Transfer Learning: An overviewTransfer Learning: An overview
Transfer Learning: An overviewjins0618
 
Deep Learning for Personalized Search and Recommender Systems
Deep Learning for Personalized Search and Recommender SystemsDeep Learning for Personalized Search and Recommender Systems
Deep Learning for Personalized Search and Recommender SystemsBenjamin Le
 

What's hot (20)

KERAS Python Tutorial
KERAS Python TutorialKERAS Python Tutorial
KERAS Python Tutorial
 
Introduction to Keras
Introduction to KerasIntroduction to Keras
Introduction to Keras
 
Deep Learning Made Easy with Deep Features
Deep Learning Made Easy with Deep FeaturesDeep Learning Made Easy with Deep Features
Deep Learning Made Easy with Deep Features
 
AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...
 
Deep learning
Deep learningDeep learning
Deep learning
 
Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015
Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015
Introduction to Deep Learning for Image Analysis at Strata NYC, Sep 2015
 
Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017
Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017
Getting Started with Keras and TensorFlow - StampedeCon AI Summit 2017
 
Artificial Intelligence, Machine Learning and Deep Learning
Artificial Intelligence, Machine Learning and Deep LearningArtificial Intelligence, Machine Learning and Deep Learning
Artificial Intelligence, Machine Learning and Deep Learning
 
Deep Learning: Chapter 11 Practical Methodology
Deep Learning: Chapter 11 Practical MethodologyDeep Learning: Chapter 11 Practical Methodology
Deep Learning: Chapter 11 Practical Methodology
 
AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...AI powered emotion recognition: From Inception to Production - Global AI Conf...
AI powered emotion recognition: From Inception to Production - Global AI Conf...
 
Introduction to deep learning
Introduction to deep learningIntroduction to deep learning
Introduction to deep learning
 
Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...
Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...
Using Deep Learning to do Real-Time Scoring in Practical Applications - 2015-...
 
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 Separating Hype from Reality in Deep Learning with Sameer Farooqui Separating Hype from Reality in Deep Learning with Sameer Farooqui
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 
Intro to Mahout -- DC Hadoop
Intro to Mahout -- DC HadoopIntro to Mahout -- DC Hadoop
Intro to Mahout -- DC Hadoop
 
Fundamental of deep learning
Fundamental of deep learningFundamental of deep learning
Fundamental of deep learning
 
Introduction to machine learning
Introduction to machine learningIntroduction to machine learning
Introduction to machine learning
 
Artificial Intelligence and Deep Learning in Azure, CNTK and Tensorflow
Artificial Intelligence and Deep Learning in Azure, CNTK and TensorflowArtificial Intelligence and Deep Learning in Azure, CNTK and Tensorflow
Artificial Intelligence and Deep Learning in Azure, CNTK and Tensorflow
 
Convolutional Neural Networks at scale in Spark MLlib
Convolutional Neural Networks at scale in Spark MLlibConvolutional Neural Networks at scale in Spark MLlib
Convolutional Neural Networks at scale in Spark MLlib
 
Transfer Learning: An overview
Transfer Learning: An overviewTransfer Learning: An overview
Transfer Learning: An overview
 
Deep Learning for Personalized Search and Recommender Systems
Deep Learning for Personalized Search and Recommender SystemsDeep Learning for Personalized Search and Recommender Systems
Deep Learning for Personalized Search and Recommender Systems
 

Similar to Deep Learning and What's Next?

deep-learning-and-what's-next-with-Chinese-annotation
deep-learning-and-what's-next-with-Chinese-annotationdeep-learning-and-what's-next-with-Chinese-annotation
deep-learning-and-what's-next-with-Chinese-annotationTao Wang
 
2019 4-nn-and-dl-tao wang@unc-v2
2019 4-nn-and-dl-tao wang@unc-v22019 4-nn-and-dl-tao wang@unc-v2
2019 4-nn-and-dl-tao wang@unc-v2Tao Wang
 
AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)
AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)
AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)Amazon Web Services
 
Deep Learning Frameworks Using Spark on YARN by Vartika Singh
Deep Learning Frameworks Using Spark on YARN by Vartika SinghDeep Learning Frameworks Using Spark on YARN by Vartika Singh
Deep Learning Frameworks Using Spark on YARN by Vartika SinghData Con LA
 
TensorFlow 16: Building a Data Science Platform
TensorFlow 16: Building a Data Science Platform TensorFlow 16: Building a Data Science Platform
TensorFlow 16: Building a Data Science Platform Seldon
 
MDEC Data Matters Series: machine learning and Deep Learning, A Primer
MDEC Data Matters Series: machine learning and Deep Learning, A PrimerMDEC Data Matters Series: machine learning and Deep Learning, A Primer
MDEC Data Matters Series: machine learning and Deep Learning, A PrimerPoo Kuan Hoong
 
rsec2a-2016-jheaton-morning
rsec2a-2016-jheaton-morningrsec2a-2016-jheaton-morning
rsec2a-2016-jheaton-morningJeff Heaton
 
"Semantic Segmentation for Scene Understanding: Algorithms and Implementation...
"Semantic Segmentation for Scene Understanding: Algorithms and Implementation..."Semantic Segmentation for Scene Understanding: Algorithms and Implementation...
"Semantic Segmentation for Scene Understanding: Algorithms and Implementation...Edge AI and Vision Alliance
 
11_Saloni Malhotra_SummerTraining_PPT.pptx
11_Saloni Malhotra_SummerTraining_PPT.pptx11_Saloni Malhotra_SummerTraining_PPT.pptx
11_Saloni Malhotra_SummerTraining_PPT.pptxSaloniMalhotra23
 
Talk@rmit 09112017
Talk@rmit 09112017Talk@rmit 09112017
Talk@rmit 09112017Shuai Zhang
 
DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018
DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018
DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018Apache MXNet
 
Data Science, Machine Learning and Neural Networks
Data Science, Machine Learning and Neural NetworksData Science, Machine Learning and Neural Networks
Data Science, Machine Learning and Neural NetworksBICA Labs
 
Big Data & Artificial Intelligence
Big Data & Artificial IntelligenceBig Data & Artificial Intelligence
Big Data & Artificial IntelligenceZavain Dar
 
Introduction to Parallel and Distributed Computing
Introduction to Parallel and Distributed ComputingIntroduction to Parallel and Distributed Computing
Introduction to Parallel and Distributed ComputingSayed Chhattan Shah
 
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...Databricks
 
01 foundations
01 foundations01 foundations
01 foundationsankit_ppt
 
Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018
Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018 Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018
Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018 Codemotion
 
talks-afanasyev2013ndnsim-tutorial.pptx
talks-afanasyev2013ndnsim-tutorial.pptxtalks-afanasyev2013ndnsim-tutorial.pptx
talks-afanasyev2013ndnsim-tutorial.pptxhazwan30
 

Similar to Deep Learning and What's Next? (20)

deep-learning-and-what's-next-with-Chinese-annotation
deep-learning-and-what's-next-with-Chinese-annotationdeep-learning-and-what's-next-with-Chinese-annotation
deep-learning-and-what's-next-with-Chinese-annotation
 
2019 4-nn-and-dl-tao wang@unc-v2
2019 4-nn-and-dl-tao wang@unc-v22019 4-nn-and-dl-tao wang@unc-v2
2019 4-nn-and-dl-tao wang@unc-v2
 
AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)
AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)
AWS re:Invent 2016: Bringing Deep Learning to the Cloud with Amazon EC2 (CMP314)
 
Deep Learning Frameworks Using Spark on YARN by Vartika Singh
Deep Learning Frameworks Using Spark on YARN by Vartika SinghDeep Learning Frameworks Using Spark on YARN by Vartika Singh
Deep Learning Frameworks Using Spark on YARN by Vartika Singh
 
TensorFlow 16: Building a Data Science Platform
TensorFlow 16: Building a Data Science Platform TensorFlow 16: Building a Data Science Platform
TensorFlow 16: Building a Data Science Platform
 
MDEC Data Matters Series: machine learning and Deep Learning, A Primer
MDEC Data Matters Series: machine learning and Deep Learning, A PrimerMDEC Data Matters Series: machine learning and Deep Learning, A Primer
MDEC Data Matters Series: machine learning and Deep Learning, A Primer
 
rsec2a-2016-jheaton-morning
rsec2a-2016-jheaton-morningrsec2a-2016-jheaton-morning
rsec2a-2016-jheaton-morning
 
"Semantic Segmentation for Scene Understanding: Algorithms and Implementation...
"Semantic Segmentation for Scene Understanding: Algorithms and Implementation..."Semantic Segmentation for Scene Understanding: Algorithms and Implementation...
"Semantic Segmentation for Scene Understanding: Algorithms and Implementation...
 
11_Saloni Malhotra_SummerTraining_PPT.pptx
11_Saloni Malhotra_SummerTraining_PPT.pptx11_Saloni Malhotra_SummerTraining_PPT.pptx
11_Saloni Malhotra_SummerTraining_PPT.pptx
 
Talk@rmit 09112017
Talk@rmit 09112017Talk@rmit 09112017
Talk@rmit 09112017
 
Introduction to deep learning
Introduction to deep learningIntroduction to deep learning
Introduction to deep learning
 
DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018
DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018
DeepLearning001&ApacheMXNetWithSparkForInference-ACNA2018
 
Amazon Deep Learning
Amazon Deep LearningAmazon Deep Learning
Amazon Deep Learning
 
Data Science, Machine Learning and Neural Networks
Data Science, Machine Learning and Neural NetworksData Science, Machine Learning and Neural Networks
Data Science, Machine Learning and Neural Networks
 
Big Data & Artificial Intelligence
Big Data & Artificial IntelligenceBig Data & Artificial Intelligence
Big Data & Artificial Intelligence
 
Introduction to Parallel and Distributed Computing
Introduction to Parallel and Distributed ComputingIntroduction to Parallel and Distributed Computing
Introduction to Parallel and Distributed Computing
 
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...
Bringing an AI Ecosystem to the Domain Expert and Enterprise AI Developer wit...
 
01 foundations
01 foundations01 foundations
01 foundations
 
Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018
Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018 Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018
Deep learning beyond the learning - Jörg Schad - Codemotion Rome 2018
 
talks-afanasyev2013ndnsim-tutorial.pptx
talks-afanasyev2013ndnsim-tutorial.pptxtalks-afanasyev2013ndnsim-tutorial.pptx
talks-afanasyev2013ndnsim-tutorial.pptx
 

Recently uploaded

The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 3652toLead Limited
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersThousandEyes
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesSinan KOZAK
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?XfilesPro
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsMemoori
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAndikSusilo4
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptxLBM Solutions
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhisoniya singh
 

Recently uploaded (20)

The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial Buildings
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & Application
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptx
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
 

Deep Learning and What's Next?

  • 1. Copyright © SAS Institute Inc. All rights reserved. Deep Learning and What’s Next? Tao Wang (t.wang@sas.com) 2018 AI-Now
  • 2. Copyright © SAS Institute Inc. All rights reserved. Part 1: Introduction
  • 3. Copyright © SAS Institute Inc. All rights reserved. AI vs. Machine Learning Take 1 3 Source: https://towardsdatascience.com/cousins-of-artificial-intelligence-dda4edc27b55
  • 4. Copyright © SAS Institute Inc. All rights reserved. AI vs. Machine Learning Take 2 Machine learning is a “field of study that gives computers the ability to learn without being explicitly programmed.” – Arthur Samuel, 1959 4 “Artificial intelligence (AI), sometimes called machine intelligence, is intelligence demonstrated by machines.” – Wikipedia, retrieved 2018
  • 5. Copyright © SAS Institute Inc. All rights reserved. AI vs. Machine Learning 5 Take 3
  • 6. Copyright © SAS Institute Inc. All rights reserved. AI vs. Machine Learning Take 4 – my own version 6 AI: goal Analytics: business Machine Learning: means
  • 7. Copyright © SAS Institute Inc. All rights reserved. Styles of Machine Learning Other styles from different perspectives (active learning, transfer learning, multi-task learning, adversarial learning, …) Supervised Semi-SupervisedUnsupervised Reinforcement • Training data has labelled target • Predict label for unseen data • Labels are known for a subset of data • A blend of supervised and unsupervised learning • Labels unknown • Find patterns and gain insights from the data • An agent selects actions to maximize reward in an environment • Face detection, fraud detection, patient identification • Customer clustering • Association rule mining • Pre-processing for supervised learning to reduce labelling cost and enhance accuracy • Game AI • Robotics 7 Source: [5] X. Hunt, et al.
  • 8. Copyright © SAS Institute Inc. All rights reserved. What Can Machine Learning Do? And so many other things! Prediction Decision and Policy-Making Data Exploration Rule Learning • Classification • Regression • Clustering • Dimension reduction • Anomaly detection • Feature engineering • Identifying relational rules within data • Association rule mining • Supervised learning • Semi-supervised learning • Often unsupervised learning • Also supervised and semi-supervised learning • Reinforcement learning • Supervised learning • Unsupervised learning • Semi-supervised learning • Learning through trial and error to identify best action • Game playing • Control problems 8 Source: [5] X. Hunt, et al.
  • 9. Copyright © SAS Institute Inc. All rights reserved. 9 Machine Learning and Deep Learning Machine Learning Deep Learning AlphaGo Bidirectional Encode AlphaFold Image sources: Siliconangle, googleblog, profacgen
  • 10. Copyright © SAS Institute Inc. All rights reserved. Part 2: Deep Learning = Deep Neural Networks (DL = DNN)? 10
  • 11. Copyright © SAS Institute Inc. All rights reserved. 11 Deep Learning: Model with Depth Shallow Deep Learning • Model with one or a few layers • Multiple layers, layer-by-layer processing • Feature extraction/transformation • Learn complex structures Data Model Output Data Output Model Layer Layer Layer Deep Learning (DL) = Deep Neural Networks (DNN), ignoring subtle stuff Source: [5] X. Hunt, et al.
  • 12. Copyright © SAS Institute Inc. All rights reserved. Pros and cons • Advantages 1. Requires minimal feature engineering (end-to-end ML) 2. Flexible structures 3. Learning often improves with more data 4. Proven track records in speech/text processing and image/video recognition • Disadvantages 1. Difficult to interpret – often treated as a “black-box” model 2. Long training time, over-fitting 3. Hard to train, non-repeatable results, numerous architectures/hyper-parameters 4. Requires a large amount of training data to get good models 12
  • 13. Copyright © SAS Institute Inc. All rights reserved. Why so popular? 1. End2end/distributed feature learning 2. Advances in algorithms/optimizations (min-batch, drop-out, BN, SGD, etc.) 3. Cloud computing and GPU made it possible to train very deep models 4. Proven track records in speech/text processing and image/video recognition 13 Source: [6] D. Silver
  • 14. Copyright © SAS Institute Inc. All rights reserved. More about DNN • When should I use DNN? • Deal with image/video/text/speech • Works for small-medium data, but prefers big data • The underlying model is complex and non-linear • OK with non-interpretability, and/or have cloud/GPU • Common DNN architectures • Deep Forward Nets • Convolutional neural networks (CNN) • Recurrent neural networks (RNN) • Stacked auto-encoders 14
  • 15. Copyright © SAS Institute Inc. All rights reserved. Deep Forward Net • A flat architecture • Regression and classification DNN architectures 1 15 Source: [4] W. Thompson
  • 16. Copyright © SAS Institute Inc. All rights reserved. Convolutional neural network (CNN) • A feedforward neural net with conv layers • 3D volumes of neurons • Feature extraction • Applications: image/video recognition, NLP DNN architectures 2 16 Source: [4] W. Thompson
  • 17. Copyright © SAS Institute Inc. All rights reserved. Recurrent neural network (RNN) • Contain at least one feed-back connection • Time-series forecasting, speech recognition DNN architectures 3 delay h1(t)h1(t-1) 17 Source: [4] W. Thompson
  • 18. Copyright © SAS Institute Inc. All rights reserved. Auto-encoder • A generative graphical model • Feature coding, dimension reduction and compression DNN architectures 4 18 Source: [4] W. Thompson
  • 19. Copyright © SAS Institute Inc. All rights reserved. DNN supported by SAS 19 Source: [7] White paper: How to Do Deep Learning With SAS?
  • 20. Copyright © SAS Institute Inc. All rights reserved. SAS platform for DL 20
  • 21. Copyright © SAS Institute Inc. All rights reserved. SAS® Visual Machine learning and Machine Learning (VDMML) Visual “drag & drop” GUI 21
  • 22. Copyright © SAS Institute Inc. All rights reserved. Applications of SAS Deep Learning 22 Source: [7] White paper: How to Do Deep Learning With SAS?
  • 23. Copyright © SAS Institute Inc. All rights reserved. Applications Input DNN Military Surveillance Speech recognition Fraud Detection Image classification Autonomous Vehicles Patient Identification 23 Source: [4] W. Thompson
  • 24. Copyright © SAS Institute Inc. All rights reserved. Autonomous vehicles An application of DNN The tipping point: level 3 Partial Autonomy Source: https://iq.intel.com/autonomous-cars-road-ahead/ Expected Timeline for Full Autonomy? Source: https://thelastdriverlicenseholder.com/2016/12/29/expected-timeline-for-full-autonomy/ Focus on Level 3 and deliver! 24
  • 25. Copyright © SAS Institute Inc. All rights reserved. Navigant Research Leaderboard Automated Driving Vehicles Source: https://www.navigantresearch.com/research/navigant-research-leaderboard-automated-driving-vehicles 25
  • 26. Copyright © SAS Institute Inc. All rights reserved. End to End Learning for Self-Driving Cars Finally, you are talking about a paper… • arXiv:1604.07316, Apr 2016, from NVIDIA • Basic idea: behavioral cloning, train the car to drive like you do • Uses CNN to map images from cameras to steering commands • Never explicitly train the CNN to detect/follow lanes, path planning, etc. 26 High-level view of the data collection system Training the CNN Self-driving Source: [1] M. Bojarski, et al.
  • 27. Copyright © SAS Institute Inc. All rights reserved. CNN architecture and the core source code 27 Read it from bottomup. Input layer, normalization layer, 5 conv2D layers: feature extraction. 3 fully-connected layers, output: controller. 27M connections, 250K parameters, 3MB in size. Source: arXiv:1604.07316 Source: github, the NVIDIA 2016 paper implementation
  • 28. Copyright © SAS Institute Inc. All rights reserved. Part 3: What’s next? 28 THE POWER OF THE PACK AI with THE POWER OF DIVERSITY AI with THE POWER OF TRUST AI with
  • 29. Copyright © SAS Institute Inc. All rights reserved. 29 Rediscover Deep Learning End to End 1 Distributed Feature Learning 2 Big Data Big Model 3
  • 30. Copyright © SAS Institute Inc. All rights reserved. 30 Source: Yoshua Bengio Source: Pablo Picasso Capsule Networks – power of the pack Source: CB Insights, State of AI Source: Forbes
  • 31. Copyright © SAS Institute Inc. All rights reserved. Capsule Network paper Yep, talking about paper again • S. Sabour, N. Frosst, G. Hinton, Dynamic Routing Between Capsules, Google Brain, NIPS 2017, https://arxiv.org/abs/1710.09829 • Introduced years ago by Hinton, but was not working properly until now • Widely considered as the beginning of a new chapter of deep learning • Some follow-up papers, such as Matrix Capsules With EM Routing • https://openreview.net/pdf?id=HJWLfGWRb, ICLR 2018 • Introduced capsule convolution layer and more sophisticated routing 31 Source: http://www.cs.toronto.edu/~hinton
  • 32. Copyright © SAS Institute Inc. All rights reserved. Dynamic Routing Between Capsules • Idea #1: capsule is an encapsulated vector/matrix in the network • A capsule is a group of neurons that represents the parameters of some specific feature. • A vector or matrix is extended from a scalar • The length represents the probability of the presence of a feature or an object • Each dimension within the capsule represents the detailed information of location, size, orientation, etc. • Idea #2: routing by agreement • Lower-level capsule (which is near input) prefers to send its output to higher-level (which is near output) capsules with “similar” prediction • Cosine similarity is used to measure the agreement 32
  • 33. Copyright © SAS Institute Inc. All rights reserved. CapsNet Architecture ▪ Input: MNIST dataset ▪ ReLU conv1: extracts local features ▪ PrimaryCaps: forms new neural unit (capsule) ▪ DigitCaps: contains 10 capsules (number 0 to 9) ▪ Cosine similarity (routing) is applied between PrimaryCaps and DigitCaps ▪ Reconstruction: a regularization method to encourage the capsules to encode the input digit Figure 1: A simple CapsNet with 3 layers Figure 2: Reconstruct a digit from the DigitCaps layer representation source: https://arxiv.org/abs/1710.09829 33
  • 34. Copyright © SAS Institute Inc. All rights reserved. Core source code Source: github, the NIPS 2017 paper implementation 34
  • 35. Copyright © SAS Institute Inc. All rights reserved. Numerical results of the NIPS paper source: https://arxiv.org/abs/1710.09829 35
  • 36. Copyright © SAS Institute Inc. All rights reserved. 36 𝐸 = 𝐸 − 𝐷 Deep Forest – power of diversity
  • 37. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper series • Deep Forest [10], using RF to do DL with the “3 key ingredients”: • In-model feature extraction and transformation, end-to-end machine learning • Layer by layer processing, distributed representation learning • Complex model • AutoEncoder by Forest [11] • The first tree ensemble based auto-encoder • Multi-Layered Gradient Boosting Decision Trees [12] • A variant of target propagation, pseudo-mapping F, pseudo-inverse-mapping G, pseudo-label Z (F-G-Z framework) • More to come? 37 Why always Neural Nets? We can do DL using Decision Trees!
  • 38. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Can you stop talking about papers? No, papers are fun, ☺ • IJCAI 2017 paper [10], by Zhou and his student • DeepForest = Forest ensemble, double-happiness (ensemble of ensembles) 1. Multi-grain scanning, sliding window to extract features 2. Cascade of multiple random forests layers, for prediction • Very few hyper-parameters (how nice!) & as good as DNN • Default settings are good for many applications • Non-differentiable model, no back propagation 38 Source: https://en.wikipedia.org/wiki/Zhi-Hua_Zhou
  • 39. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Problems of DNN • Too many hyper-parameters (like an art rather than science) • Does not work well for small data • Model architecture/complexity is determined in advance (via tuning) • Often overly complicated • Shortcut connection, pruning, binarization, etc. are often applied 39
  • 40. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Why deep forest? Motivations? • Decision trees • Architecture learning (grow/split until done) • Data driven • Almost unbeatable on tabular data in Kaggle • Motivations • DL = DNN? • Can we do DL with non-differentiable models (no back-propagation)? • Maybe repeatable results (think SGD)? 40
  • 41. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Inspiration from DNN • Distributed representation learning (end to end, in-model feature trans.) • Layer-by-layer processing • Model complexity 41 Source: [10] Z. Zhou and J. Feng
  • 42. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Multi-Grained Scanning for Feature Engineering 42 • Sequential relationships are important • Spatial relationships are important Source: [10] Z. Zhou and J. Feng
  • 43. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Cascade Forest Structure for Prediction • Ensemble of ensembles • K-fold cross validation • Architecture learning (stop growing when satisfied) 43 Source: [10] Z. Zhou and J. Feng
  • 44. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Class Vector Generation 44 Source: [10] Z. Zhou and J. Feng
  • 45. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Overall Architecture 45 Source: [10] Z. Zhou and J. Feng
  • 46. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Hyper-parameters and default settings 46 Source: [10] Z. Zhou and J. Feng
  • 47. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Experimental results 47 Image Categorization Face Recognition Music Classification Hand Movement Recognition Source: [10] Z. Zhou and J. Feng
  • 48. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper More experimental results 48 Sentiment Classification Low-Dimensional Data High-Dimensional Data (hard to beat successful method at its killer-app with a brand-new algorithm) Source: [10] Z. Zhou and J. Feng
  • 49. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Running time • PC with 2 Intel E5 2695 v4 CPUs (18 cores) • IMDB dataset (25,000 examples, with 5,000 features) • Deep Forest: 40 minutes • DNN: can take over 60 minutes 49
  • 50. Copyright © SAS Institute Inc. All rights reserved. Deep Forest paper Hyper-parameter sensitivity 50 Source: [10] Z. Zhou and J. Feng
  • 51. Copyright © SAS Institute Inc. All rights reserved. 51 AI Analytics Machine Learning Blockchain – power of trust Source: pixabay
  • 52. Copyright © SAS Institute Inc. All rights reserved. A Unified Analytical Framework for Trustable Machine Learning and Automation Running with Blockchain 52 Source: [14] T. Wang
  • 53. Copyright © SAS Institute Inc. All rights reserved. Further reading list • Reinforcement learning (play, explore, control, interact) • An agent selects actions to maximize reward in an environment • AI = Deep RL (D. Silver, 2016) vs. RL does not really work (I. Goodfellow, 2018) • Generative adversarial networks (GAN) [9] • Unsupervised learning using supervised learning as sampling model • Infers models in a competing game with Generator (G) and Discriminator (D) • Provides an attractive alternative to maximum likelihood techniques. • Y. LeCun: “…There are many interesting development in deep learning…The most important one, …, is adversarial training….” • Adaptive Neural Trees (ANT), https://arxiv.org/abs/1807.06699, 2018 - NN: end2end/distributed representation learning with pre-specified architecture, image/sequence - DT: architecture learning with pre-specified features, tabular data • BERT – Bidirectional Encoding model • AlphaFold 53
  • 54. Copyright © SAS Institute Inc. All rights reserved. AI winter is coming? 54 Source: https://blog.piekniewski.info/2018/05/28/ai-winter-is-well-on-its-way/ Source: Google trend
  • 55. Copyright © SAS Institute Inc. All rights reserved. Closing Remarks AI and machine learning are very hard – just keep trying! 55
  • 56. Copyright © SAS Institute Inc. All rights reserved. Selected References • [1] M. Bojarski, et al., End to End Learning for Self-Driving Cars, arXiv:1604.07316, 2016. • [2] S. Sabour, N. Frosst, G. Hinton, Dynamic Routing Between Capsules, Google Brain, NIPS 2017, https://arxiv.org/abs/1710.09829 • [3] D. Silver, A. Huang, et, al. (2016). "Mastering the game of Go with deep neural networks and tree search". Nature 529 (7587): 484– 489. • [4] W. Thompson, et al., Introduction to Deep learning, SAS, 2016. • [5] X. Hunt, et al., Machine Learning Landscape, SAS, 2017. • [6] D. Silver, Tutorial: Deep Reinforcement Learning, 2017. • [7] White paper: How to Do Deep Learning With SAS? 2018. • [8] Y. LeCun, et al., Deep learning, Nature, 2015. • [9] I. Goodfellow, et al., Generative Adversarial Net, https://arxiv.org/abs/1406.2661 • [10] Z. Zhou and J. Feng, Deep Forest, IJCAI 2017. • [11] J. Feng and Z. Zhou, AutoEncoder by Forest, AAAI 2018. • [12] J. Feng, Y. Yu, Z. Zhou, Multi-Layered Gradient Boosting Decision Trees, https://arxiv.org/abs/1806.00007, 2018 • [13] R. Tanno, et al., Adaptive Neural Trees, https://arxiv.org/abs/1807.06699, 17 Jul 2018. • [14] T. Wang, A Unified Analytical Framework for Trustable Machine Learning and Automation Running with Blockchain, IEEE Big Data Workshops, 2018. 56
  • 57. Copyright © SAS Institute Inc. All rights reserved. Running for 2019 ACM SIGAI Vice-Chair Shameless election/campaign ads Vote for Tao Wang 57
  • 58. Copyright © SAS Institute Inc. All rights reserved. Guest Speaker Dr. Kirk Borne 58 Source: https://twitter.com/KirkDBorne