artificial intelligence. what machines can learn and how to … · 2017-05-24 · artificial...
TRANSCRIPT
Artificial Intelligence. What machines can learn and how to implement Machine Learning
project in you organization?
Antons Mislēvičs
1. Over 9 years @ Microsoft Over 12 years of SharePoint experience
2. Projects: Portals, ECM, Social, Search, BPMS – Architect, Lead Developer
Architecture Design Sessions – Subject Matter Expert on SharePoint
Cloud Solutions: Office 365, SharePoint Online, Windows Azure, SharePoint / Office 365 Dev Patterns & Practices (PnP) Core Team member
3. Countries: Latvia, Lithuania, Estonia, Ukraine, Kazakhstan, Malta, Cyprus, Switzerland,
UK, Germany, France, Russia, Brunei, Azerbaijan, Georgia, Sweden, USA
4. PhD in Engineering of Computer Systems: Thesis: “Mobile Agents for Business Process Management Support in Cloud
Computing Environments”
5. Lead Researcher at Riga Technical University Department of Artificial Intelligence and Systems Engineering
Antons Mislēvičs
Senior Consultant
Microsoft Services EMEA
Agenda
1. What machines can do today?
2. How Machine Learning works?
3. How to implement Machine Learning in your project?
4. How the future looks like?
Chess: IBM Deep Blue 3 ½ - Gary Kasparov 2 ½ (1997)
Kasparov vs Deep Blue the rematch, 1997
https://www.research.ibm.com/deepblue/
Jeopardy: IBM Watson beats champions (2011)
Final Jeopardy! and the Future of Watson
http://www-03.ibm.com/marketing/br/watson/what-is-watson/the-future-of-watson.html
Go: Google AlphaGo 4 – Lee Sedol 1 (2016)
AlphaGo
https://deepmind.com/alpha-go
Poker: Libratus and DeepStack beat top pros (2017)
Carnegie Mellon Artificial Intelligence Beats Top Poker Pros: https://www.cmu.edu/news/stories/archives/2017/january/AI-beats-poker-pros.html
Brains Vs. AI Rematch: Why Poker?: https://www.youtube.com/watch?v=JtyA2aUj4WI
Tough poker player: Brains Vs. AI update: https://www.youtube.com/watch?v=CRiH8yCskAE
Safe and Nested Endgame Solving for Imperfect-Information Games, N. Brown, T. Sandholm, 2017: http://www.cs.cmu.edu/~sandholm/safeAndNested.aaa17WS.pdf
DeepStack: Expert-Level Artificial Intelligence in No-Limit Poker, 2017: https://arxiv.org/abs/1701.01724
Image RecognitionError rates – human vs machine
1. Traffic Sign Recognition (IJCNN 2011):
– Human: 1.16%
– Machine: 0.54%
2. Handwritten Digits (MNIST):
– Human: approx. 0.2%
– Machine: 0.23% (2012)
The German Traffic Sign Recognition Benchmark: http://benchmark.ini.rub.de/?section=gtsrb&subsection=results
THE MNIST DATABASE of handwritten digits: http://yann.lecun.com/exdb/mnist/
Large Scale Visual Recognition Challenge (ILSVRC)
2015 challenge:
– Object detection - 200 categories
– Object recognition – 1000 categories
– Object detection from video – 30 categories
– Scene classification – 401 categories
Large Scale Visual Recognition Challenge 2015 (ILSVRC2015)
http://image-net.org/challenges/LSVRC/2015/index#maincomp
Large Scale Visual Recognition Challenge 2015 – Results: http://image-net.org/challenges/LSVRC/2015/results
Microsoft Researchers’ Algorithm Sets ImageNet Challenge Milestone, 2015: https://www.microsoft.com/en-us/research/microsoft-researchers-algorithm-sets-imagenet-challenge-milestone/
Microsoft Research Team:
“To our knowledge, our result is the first to surpass human-
level performance…on this visual recognition challenge”
Machines can understand the meaning…
Show and Tell: A Neural Image Caption Generator, O. Vinyals, A. Toshev, S. Bengio, D. Erhan, 2015: http://arxiv.org/abs/1411.4555v2
DenseCap: Fully Convolutional Localization Networks for Dense Captioning, J. Johnson, A. Karpathy, L. Fei-Fei, 2015: http://arxiv.org/abs/1511.07571
Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models, R. Kiros, R. Salakhutdinov, R. S. Zemel, 2014: http://arxiv.org/abs/1411.2539
Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models, R. Kiros, R. Salakhutdinov, R. S. Zemel, 2014: http://arxiv.org/abs/1411.2539
Self-Driving Cars
Google Self-Driving Car Project - How it drives: https://www.google.com/selfdrivingcar/how/
Autopilot Full Self-Driving Hardware (Neighborhood Long), Tesla Motors: https://vimeo.com/192179727
2. The Next Rembrandt [4]
1. A Neural Algorithm of Artistic Style, 2015: https://arxiv.org/abs/1508.06576
2. Supercharging Style Transfer, 2016: https://research.googleblog.com/2016/10/supercharging-style-transfer.html
3. Neural Doodle:, 2016: https://github.com/alexjc/neural-doodle
4. The Next Rembrandt: https://www.nextrembrandt.com/
5. Image Completion with Deep Learning in TensorFlow, 2016: https://bamos.github.io/2016/08/09/deep-completion/
6. Neural Enhance, 2016: https://github.com/alexjc/neural-enhance
Machines get creative…1. Reproduce artistic style [1, 2, 3]
4. Enhance images [6]
3. Complete images [5]
1. WaveNet: A Generative Model for Raw Audio: https://deepmind.com/blog/wavenet-generative-model-raw-audio/
2. WaveNet: A Generative Model for Raw Audio, 2016: https://arxiv.org/abs/1609.03499
3. Historic Achievement: Microsoft researchers reach human parity in conversational speech recognition: https://blogs.microsoft.com/next/2016/10/18/historic-achievement-microsoft-researchers-reach-human-parity-conversational-speech-recognition/
4. Achieving Human Parity in Conversational Speech Recognition, 2016: http://arxiv.org/abs/1610.05256
Text to speech and voice recognition…
2. Recognize voice [3, 4]
1. Talk - text to speech (WaveNet) [1, 2]
2. Generate handwriting [2, 3]
3. Translate texts [4, 5]
1. Composing Music With Recurrent Neural Networks: http://www.hexahedria.com/2015/08/03/composing-music-with-recurrent-neural-networks/
2. Generating Sequences With Recurrent Neural Networks, A. Graves, 2014: http://arxiv.org/abs/1308.0850
3. Alex Graves’s RNN handwriting generation demo: http://www.cs.toronto.edu/~graves/handwriting.html
4. University of Montreal, Lisa Lab, Neural Machine Translation demo: http://lisa.iro.umontreal.ca/mt-demo
5. Fully Character-Level Neural Machine Translation without Explicit Segmentation, J.Lee, K. Cho, T. Hofmann, 2016: http://arxiv.org/abs/1610.03017
What else machines can do?
1. Compose music [1]
Machine Learning process
Introducing Azure Machine Learning, D. Chappell, 2015:
http://www.davidchappell.com/writing/white_papers/Introducing-Azure-ML-v1.0--Chappell.pdf
Machine Learning questions
1. How much / how many? Regression
2. Which category? Classification
3. Which groups? Clustering
4. Is it weird? Anomaly Detection
5. Which action? Reinforcement Learning
Data Science for Rest of Us, B. Rohrer, 2015:
https://channel9.msdn.com/blogs/Cloud-and-Enterprise-Premium/Data-Science-for-Rest-of-Us
Housing prices by square feetPrice Square Feet
125,999 950
207,190 1125
227,555 1400
319,010 1750
345,846 1525
350,000 1690
437,301 2120
450,999 2500
605,000 3010
641,370 3250
824,280 3600
1,092,640 3700
1,187,550 4500
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
1,100,000
1,200,000
Input variable/FeatureOutput variable
Housing prices hypothesisPrice Square Feet
125,999 950
207,190 1125
227,555 1400
319,010 1750
345,846 1525
350,000 1690
437,301 2120
450,999 2500
605,000 3010
641,370 3250
824,280 3600
1,092,640 3700
1,187,550 4500
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
1,100,000
1,200,000
Hypothesis
Using model to predict house pricePrice Square Feet
125,999 950
207,190 1125
227,555 1400
319,010 1750
345,846 1525
350,000 1690
437,301 2120
450,999 2500
??? 2700
605,000 3010
641,370 3250
824,280 3600
1,092,640 3700
1,187,550 4500
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
1,100,000
1,200,000
Prediction “errors” & improving modelsPrice Square Feet
125,999 950
207,190 1125
227,555 1400
319,010 1750
345,846 1525
350,000 1690
437,301 2120
450,999 2500
??? 2700
605,000 3010
641,370 3250
824,280 3600
1,092,640 3700
1,187,550 4500
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
1,100,000
1,200,000
Cost function (sq. error function)
Try different algorithmPrice Square Feet
125,999 950
207,190 1125
227,555 1400
319,010 1750
345,846 1525
350,000 1690
437,301 2120
450,999 2500
??? 2700
605,000 3010
641,370 3250
824,280 3600
1,092,640 3700
1,187,550 4500
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
1,100,000
1,200,000
Get more data
0
1000
2000
3000
4000
5000
6000
0 100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
900,000
1,000,000
1,100,000
1,200,000
1,300,000
1,400,000
1,500,000
1,600,000
1,700,000
1,800,000
1,900,000
2,000,000
2,100,000
Use more data variables (features)Price Square Feet # Bedrooms # Bathrooms Fireplaces Garage Size Floors
125,999 950 1 1 0 0 1
207,190 1125 1 1 0 1 1
227,555 1400 2 1.5 1 2 1
319,010 1750 2 1.5 0 2 2
345,846 1525 3 2 1 2 1
350,000 1690 3 1.5 1 2 1.5
437,301 2120 3 2.5 2 3 2
450,999 2500 3 2.5 1 2 1.5
605,000 3010 4 2.5 2 3 2
641,370 3250 3 3 1 3 2
824,280 3600 3 3 2 3 2
1,092,640 3700 5 4.5 2 3 2
1,187,550 4500 6 6 4 5 2
Years
driving Age Class
5 65 1
7 70 1
2 68 1
25 45 2
25 55 2
20 50 2
5 25 1
3 22 1
8 30 1
15 35 2
… … …
12 38 ???
Classification – 2 classes
Hypothesis / Classifier
Years driving
Age
Input variables/Features Output variable
1. Input variables / Features:
– Years driving
– Age
2. Output variable:
– Class: Yellow, Green, Blue
3. One-vs-rest approach:
– Train classifier for each class
– Select class that returned highest confidence score
Classification – more than 2 classes
Classifier 3
Years driving
Age
Classifier 2
Classifier 1
1. Define business value and
how it can be measured
- How much / how many?
- Which category?
- Which groups?
- Is it weird?
- Which action?
3. Build models:
- “Black-box” – pure statistical analysis
of large amounts of data
- ”Soft-box” – heuristic insights from the
knowledge of experts
4. Integrate into production systems
- Adjust business processes
- Redesign existing systems
5. Drive adoption!
Implementing Machine Learning project2. Prepare data:
- Internal data sources
- External data sources
Cortana Intelligence Suite services
Cortana Intelligence Suite: https://azure.microsoft.com/en-us/suites/cortana-intelligence-suite/
Typical Machine Learning scenarios
Customers
– Recommendations
– Customer Churn
– Customer Segmentation
Operations
– Predictive Maintenance
– Anomaly Detection
– Optimization
Security & Risk
– Credit Risk
– Fraud Detection
– Predict Security Threat
Cortana Intelligence Gallery: https://gallery.cortanaintelligence.com/
Cortana Intelligence Gallery - Industries: https://gallery.cortanaintelligence.com/industries/
Cortana Intelligence Gallery – Solutions: https://gallery.cortanaintelligence.com/solutions
Microsoft Cognitive Services: https://www.microsoft.com/cognitive-services
Recognize Emotions
Face Detection
Face Verification
Similar Face Searching Face Grouping
Microsoft Cognitive Services – QnA Maker: https://www.microsoft.com/cognitive-services/en-us/qnamaker
QnA Maker: https://qnamaker.ai/
Calculate cost savings and ROI
1. Generate confusion matrix for model
2. Calculate costs for each state
3. Choose the best model and
cut-off point
Cost of door maint proc/part replacement = 200 EUR
Cost of door maint proc/part replacement = 200 EUR
Cost of door repair = 600 EUR
Cost of cancelling door maint proc = - 150 EUR cost savings
A DARPA Perspective on Artificial Intelligence: https://www.youtube.com/watch?v=-O01G3tSYpU
1. Wave 1: Handcrafted Knowledge
2. Wave 2: Statistical Learning
3. Wave 3: ...
AI history
Why Machine Learning is developing rapidly?
1. Data
2. Algorithms – same approach in different domains (deep neural networks)
3. Computing power – cloud and GPUs
Introducing GeForce GTX TITAN Z: Ultimate Power, May 2014.
http://www.geforce.com/whats-new/articles/introducing-nvidia-geforce-gtx-titan-z
Deep neural networks can be fooled…
Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images, 2015: https://arxiv.org/abs/1412.1897
Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images, 2015: https://arxiv.org/abs/1412.1897
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images: http://www.evolvingai.org/fooling
Machines do not always “understand” images…
Accelerating innovation and powering new experiences with AI: https://code.facebook.com/posts/310100219388873/accelerating-innovation-and-powering-new-experiences-with-ai/
CS231n: Convolutional Neural Networks for Visual Recognition, Lecture 10: Recurrent Neural Networks: http://cs231n.stanford.edu/slides/winter1516_lecture10.pdf
Machines do not “understand” how the world works…
Accelerating innovation and powering new experiences with AI: https://code.facebook.com/posts/310100219388873/accelerating-innovation-and-powering-new-experiences-with-ai/
RI Seminar: Yann LeCun : The Next Frontier in AI: Unsupervised Learning: https://www.youtube.com/watch?v=IbjF5VjniVE
A DARPA Perspective on Artificial Intelligence: https://www.youtube.com/watch?v=-O01G3tSYpU
1. Wave 1: Handcrafted Knowledge
2. Wave 2: Statistical Learning
3. Wave 3: Contextual Adaptation
AI future
RI Seminar: Yann LeCun : The Next Frontier in AI: Unsupervised Learning: https://www.youtube.com/watch?v=IbjF5VjniVE
1. Machines need to learn/understand how the world works (get
common sense)
2. Machines need to learn a very large amount of background
knowledge (through observation and action)
3. Unsupervised learning
The future…
Playing Atari with Deep Reinforcement Learning, 2013: https://arxiv.org/abs/1312.5602
Google DeepMind's Deep Q-learning playing Atari Breakout: https://www.youtube.com/watch?v=V1eYniJ0Rnk
Playing Atari
Key takeaways
1. Imagine how ML can change your business?
2. Describe business stakeholders what ML can do
3. Start small – look for quick ML wins
– Make goals measurable – what is min required prediction accuracy?
– Focus on implementing models in production
4. Create data culture in your organization
– Improve data quality
– Invest in people
5. Do not sell ML as just “another cool thing” – define business value and calculate ROI
Slides: https://1drv.ms/f/s!AsXSX3Q3cMlAjfg_NxS_CCX6XmRBLg