SlideShare uma empresa Scribd logo
1 de 28
Baixar para ler offline
Dr. Christian Betz
Debugging AI
Data & AI Craftsmanship
2Photo by Gratisography from Pexels
„Debugging AI“ is deliberately
ambiguous.
What drives the current AI hype?
3
4
Compute Power
https://www.youtube.com/watch?v=0ibVhtuQkZA
5
More data
Photo by Negative Space from Pexels
6
New business opportunities

(plus attention)
By Dllu - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=63450446
7
and language processing.
28 Chapter 2 – Why is AI important?
AI is important because, for the first time, traditionally human
capabilities can be undertaken in software inexpensively
and at scale. AI can be applied to every sector to enable
new possibilities and efficiencies.
40 Chapter 3 – Why has AI come of age?
Specialised hardware, availability of training data, new
algorithms and increased investment, among other factors,
have enabled an inflection point in AI capability. After seven
false dawns since the 1950s, AI technology has come of age.
constraints of human experience
82 Chapter 6 – The war
While demand for AI professiona
winners and losers are emerging
Part 3: The AI Disrup
96 Chapter 7 – Europe’s
The landscape for entrepreneurs
AI startups are maturing, bringing
industries, and navigating new op
While the UK is the powerhouse o
France may extend their influence
https://www.mmcventures.com/wp-content/uploads/2019/02/The-State-of-AI-2019-Divergence.pdf
Speed of development
https://www.mmcventures.com/wp-content/uploads/2019/02/The-State-of-AI-2019-Divergence.pdf 8
years (Fig. 23), to an estimated $15bn in 2018 (CB Insights,
MMC Ventures).
Today’s leading technology companies – including Apple,
Amazon, Facebook, Google, IBM, Microsoft and Salesforce
– are also spending heavily on research and personnel to
develop and deploy AI. Internal corporate investment on AI,
among just the top 35 high tech and advanced manufacturing
companies investing in AI, may be 2.0x to 4.5x greater than the
capital invested by venture capital firms, private equity firms
and other sources of external funding combined (McKinsey),
further catalysing progress.
have increased fifteen-fold
in five years, to an estimated
$15bn in 2018.
(CB Insights, MMC Ventures)
Source: CB Insights, MMC Ventures
Fig 23. Venture capital investment in AI has increased 15-fold in five years
0
200
400
600
800
1000
1200
0
2
4
6
8
10
12
14
16
20132012 2014 2015 2016 2017 2018E
Fig. X: Venture capital investment in AI has increased 15-fold in five years
AIdeals
Disclosed Funding (right axis)
Number (left axis)
AIdealinvestment($billion)
High valuation
9
Fear
Photo by samer daboul from Pexels
It’s important to understand…
10
that these factors impact your AI project.
What is AI?
Strong AI vs. Weak AI
AI is used as a generic term for a set of tools to cope with a certain set of
problems.
Machine Learning is a subset of this AI-toolset: „Programming by example“.
Other subsets are knowledge representation, planning, reasoning.
AI uses probabilistic logic instead of boolean logic.
11
12
„The brown quick fox jumps over the lazy dog“
https://www.mcohen.io/2017/machine-learning-explained-in-three-easy-steps/
brown quick
Properties of AI problems
Hard to code „by hand“
• Requires non-formalized knowledge (experiential knowledge)
• Or even not yet existing knowledge
Afflicted with uncertainty (or missing information)
Changes rapidly (making it unreasonable to adopt software
manually)
13
Playing Chess, playing Go
14
f(s)
f(s)
f(s)
f(s)
f(s)
f(s)
f(s)
f(s)
1
0
0
1
15Photo by Gratisography from Pexels
digging deeper
into „debugging“
Recap: Properties of AI problems
Hard to code „by hand“
• Requires non-formalized knowledge (experiential knowledge)
• Or even not yet existing knowledge
Afflicted with uncertainty (or missing information)
Changes rapidly (making it unreasonable to adopt software
manually)
16
17
These properties make verification
hard „by design“
Recap: Properties of AI problems
Hard to code „by hand“
• Requires non-formalized knowledge (experiential knowledge)
• Or even not yet existing knowledge
18Photo by Pixabay from Pexels
Problems with unknown truth
For example: Medical classification problem
What is the correct diagnosis?
What is the correct therapy?
You won’t know (maybe until your patient either recovers or dies?)
Same is true for customer support systems. Is you customer satisfied?
You’ll probably only know by loosing him/her as a customer.
19https://towardsdatascience.com/gender-bias-word-embeddings-76d9806a0e17, Photo by rawpixel.com from Pexels
ML replicates bias in the data
Example: Conceptual similarities from word embeddings
With words typically collocated, you can ask you model for
conceptual similarities:
king - man + woman ⇾ queen
Depending on your input corpus, your model will give you
doctor - man + woman ⇾ nurse
Do you really run a sexist, racist chatbot on your website?
AI systems break fundamental patterns we
developed as an industry
• No (or very little) isolation. You need to verify and retrain the
whole system.
• Higher dimension of failure space
• Time intense training cycles (instead immediate feedback cycles)
• Non-stationary nature of ML systems
20http://ai.stanford.edu/~zayd/why-is-machine-learning-hard.html
21https://www.youtube.com/watch?v=piYnd_wYlT8
AI is prone to hacking. So know your tools!
We need …
22
… new quality management approaches
Quality management on input data
Implement quality management for your input data. Visualize, use
statistical metrics on the input data. Identify bias in the input data.
Establish panel of judges both on input data labelling and on
outcomes. Due to the non-stationary nature this is is an ongoing
task, not a closed project.
Use generated test data with known patterns, because otherwise
you won’t know if you miss whole categories.
23
Work with test sets
To test for non-binary outcomes (i.e., results with confidence level),
you need to handle test set as opposed to sets of single test
outcomes: For example test for outcome confidence distribution.
For non-stationary systems: establish test monitoring to accept
regression. „Accept new model if result is at least 95% of last
model.“
Do not only test outcomes, but implement inspection tools. E.g., in
RoboCup Simulation map agent movement paths.
24
Add a safety net
Implement multi-layer security fallbacks for subsets of your
problem (also to be used while testing), like „emergency break
systems“. Test these. Use for testing: If you need your security
fallback too often, your model may be bad.
Just in research: Add explainability, local properties by black-box
tests on the model to verify the „anchor rules" (https://
homes.cs.washington.edu/~marcotcr/aaai18.pdf)
25
© data42 GmbH
Speed / cost benefits of Machine
Learning often only apply if you accept
a non-zero failure rate
Think twice before implementing high stakes AI applications
(E.g., do NOT use AI to screen applications in HR)
26
26.02.18
© data42 GmbH
Derive new knowledge from AI
… to solve new problems.
27
26.02.18
Thank you very much,
and be curious!
xing.to/betz

Mais conteúdo relacionado

Mais procurados

AI-SDV 2021: Francisco Webber - Efficiency is the New Precision
AI-SDV 2021: Francisco Webber - Efficiency is the New PrecisionAI-SDV 2021: Francisco Webber - Efficiency is the New Precision
AI-SDV 2021: Francisco Webber - Efficiency is the New Precision
Dr. Haxel Consult
 
Artificial intelligence engineer course
Artificial intelligence engineer courseArtificial intelligence engineer course
Artificial intelligence engineer course
Ibrahim Khleifat
 

Mais procurados (20)

AI-SDV 2020: Special Hypertext Information Treatment in is Special Hypertext ...
AI-SDV 2020: Special Hypertext Information Treatment in is Special Hypertext ...AI-SDV 2020: Special Hypertext Information Treatment in is Special Hypertext ...
AI-SDV 2020: Special Hypertext Information Treatment in is Special Hypertext ...
 
Emerging engineering issues for building large scale AI systems By Srinivas P...
Emerging engineering issues for building large scale AI systems By Srinivas P...Emerging engineering issues for building large scale AI systems By Srinivas P...
Emerging engineering issues for building large scale AI systems By Srinivas P...
 
Dive into H2O: NYC
Dive into H2O: NYCDive into H2O: NYC
Dive into H2O: NYC
 
Machine Learning Project Lifecycle
Machine Learning Project LifecycleMachine Learning Project Lifecycle
Machine Learning Project Lifecycle
 
Model governance in the age of data science & AI
Model governance in the age of data science & AIModel governance in the age of data science & AI
Model governance in the age of data science & AI
 
Deep Learning, the Engine of the AI Revolution
Deep Learning, the Engine of the AI RevolutionDeep Learning, the Engine of the AI Revolution
Deep Learning, the Engine of the AI Revolution
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
 
No, you don't need to learn python
No, you don't need to learn pythonNo, you don't need to learn python
No, you don't need to learn python
 
#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...
#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...
#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...
 
AI-SDV 2021: Francisco Webber - Efficiency is the New Precision
AI-SDV 2021: Francisco Webber - Efficiency is the New PrecisionAI-SDV 2021: Francisco Webber - Efficiency is the New Precision
AI-SDV 2021: Francisco Webber - Efficiency is the New Precision
 
AI in Manufacturing & the Proposed EU Artificial Intelligence Act
AI in Manufacturing & the Proposed EU Artificial Intelligence ActAI in Manufacturing & the Proposed EU Artificial Intelligence Act
AI in Manufacturing & the Proposed EU Artificial Intelligence Act
 
Artificial intelligence engineer course
Artificial intelligence engineer courseArtificial intelligence engineer course
Artificial intelligence engineer course
 
Industrial Machine Learning (SIGKDD17)
Industrial Machine Learning (SIGKDD17)Industrial Machine Learning (SIGKDD17)
Industrial Machine Learning (SIGKDD17)
 
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
 
Artificial Intelligence in Project Management by Dr. Khaled A. Hamdy
Artificial Intelligence in Project Management by  Dr. Khaled A. HamdyArtificial Intelligence in Project Management by  Dr. Khaled A. Hamdy
Artificial Intelligence in Project Management by Dr. Khaled A. Hamdy
 
MLSEV Virtual. ML: Business Perspective
MLSEV Virtual. ML: Business PerspectiveMLSEV Virtual. ML: Business Perspective
MLSEV Virtual. ML: Business Perspective
 
AI for Software Engineering
AI for Software EngineeringAI for Software Engineering
AI for Software Engineering
 
Artificial intelligence in supply chain and manufacturing - An IBM case study...
Artificial intelligence in supply chain and manufacturing - An IBM case study...Artificial intelligence in supply chain and manufacturing - An IBM case study...
Artificial intelligence in supply chain and manufacturing - An IBM case study...
 
Latent Panelists Affinities: a Helixa case study
Latent Panelists Affinities: a Helixa case studyLatent Panelists Affinities: a Helixa case study
Latent Panelists Affinities: a Helixa case study
 
Expert System - Artificial intelligence
Expert System - Artificial intelligenceExpert System - Artificial intelligence
Expert System - Artificial intelligence
 

Semelhante a Debugging AI

The Black Box: Interpretability, Reproducibility, and Data Management
The Black Box: Interpretability, Reproducibility, and Data ManagementThe Black Box: Interpretability, Reproducibility, and Data Management
The Black Box: Interpretability, Reproducibility, and Data Management
mark madsen
 
State of AI Report 2022 - ONLINE.pptx
State of AI Report 2022 - ONLINE.pptxState of AI Report 2022 - ONLINE.pptx
State of AI Report 2022 - ONLINE.pptx
EithuThutun
 
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
James Anderson
 

Semelhante a Debugging AI (20)

Debugging AI
Debugging AIDebugging AI
Debugging AI
 
Japan 20200724 v13
Japan 20200724 v13Japan 20200724 v13
Japan 20200724 v13
 
Towards the Industrialization of AI
Towards the Industrialization of AITowards the Industrialization of AI
Towards the Industrialization of AI
 
State Of AI 2018
State Of AI 2018State Of AI 2018
State Of AI 2018
 
UNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSIS
UNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSISUNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSIS
UNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSIS
 
UNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSIS
UNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSISUNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSIS
UNCOVERING FAKE NEWS BY MEANS OF SOCIAL NETWORK ANALYSIS
 
Popular Machine Learning Myths
Popular Machine Learning Myths Popular Machine Learning Myths
Popular Machine Learning Myths
 
O'Reilly ebook: Machine Learning at Enterprise Scale | Qubole
O'Reilly ebook: Machine Learning at Enterprise Scale | QuboleO'Reilly ebook: Machine Learning at Enterprise Scale | Qubole
O'Reilly ebook: Machine Learning at Enterprise Scale | Qubole
 
EPR Annual Conference 2020 Workshop 1 - Simon Uytterhoeven
EPR Annual Conference 2020 Workshop 1 - Simon Uytterhoeven EPR Annual Conference 2020 Workshop 1 - Simon Uytterhoeven
EPR Annual Conference 2020 Workshop 1 - Simon Uytterhoeven
 
The Black Box: Interpretability, Reproducibility, and Data Management
The Black Box: Interpretability, Reproducibility, and Data ManagementThe Black Box: Interpretability, Reproducibility, and Data Management
The Black Box: Interpretability, Reproducibility, and Data Management
 
State of AI Report 2022 - ONLINE.pptx
State of AI Report 2022 - ONLINE.pptxState of AI Report 2022 - ONLINE.pptx
State of AI Report 2022 - ONLINE.pptx
 
State of AI Report 2022 - ONLINE.pdf
State of AI Report 2022 - ONLINE.pdfState of AI Report 2022 - ONLINE.pdf
State of AI Report 2022 - ONLINE.pdf
 
Quant university MRM and machine learning
Quant university MRM and machine learningQuant university MRM and machine learning
Quant university MRM and machine learning
 
CWIN17 san francisco-ai implementation-pub
CWIN17 san francisco-ai implementation-pubCWIN17 san francisco-ai implementation-pub
CWIN17 san francisco-ai implementation-pub
 
Where have all the data entry candidates gone?
Where have all the data entry candidates gone?Where have all the data entry candidates gone?
Where have all the data entry candidates gone?
 
2024_개보위_개인정보 미래포럼_의료 인공지능 모델과 프라이버시 이슈.pdf
2024_개보위_개인정보 미래포럼_의료 인공지능 모델과 프라이버시 이슈.pdf2024_개보위_개인정보 미래포럼_의료 인공지능 모델과 프라이버시 이슈.pdf
2024_개보위_개인정보 미래포럼_의료 인공지능 모델과 프라이버시 이슈.pdf
 
(In)convenient truths about applied machine learning
(In)convenient truths about applied machine learning(In)convenient truths about applied machine learning
(In)convenient truths about applied machine learning
 
Ezml Stanford 2015
Ezml Stanford 2015Ezml Stanford 2015
Ezml Stanford 2015
 
Responsible Generative AI Design Patterns
Responsible Generative AI Design PatternsResponsible Generative AI Design Patterns
Responsible Generative AI Design Patterns
 
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
GDG Cloud Southlake 28 Brad Taylor and Shawn Augenstein Old Problems in the N...
 

Último

Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
Joaquim Jorge
 

Último (20)

Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
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
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
 
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptx
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
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
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 

Debugging AI

  • 1. Dr. Christian Betz Debugging AI Data & AI Craftsmanship
  • 2. 2Photo by Gratisography from Pexels „Debugging AI“ is deliberately ambiguous.
  • 3. What drives the current AI hype? 3
  • 5. 5 More data Photo by Negative Space from Pexels
  • 6. 6 New business opportunities
 (plus attention) By Dllu - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=63450446
  • 7. 7 and language processing. 28 Chapter 2 – Why is AI important? AI is important because, for the first time, traditionally human capabilities can be undertaken in software inexpensively and at scale. AI can be applied to every sector to enable new possibilities and efficiencies. 40 Chapter 3 – Why has AI come of age? Specialised hardware, availability of training data, new algorithms and increased investment, among other factors, have enabled an inflection point in AI capability. After seven false dawns since the 1950s, AI technology has come of age. constraints of human experience 82 Chapter 6 – The war While demand for AI professiona winners and losers are emerging Part 3: The AI Disrup 96 Chapter 7 – Europe’s The landscape for entrepreneurs AI startups are maturing, bringing industries, and navigating new op While the UK is the powerhouse o France may extend their influence https://www.mmcventures.com/wp-content/uploads/2019/02/The-State-of-AI-2019-Divergence.pdf Speed of development
  • 8. https://www.mmcventures.com/wp-content/uploads/2019/02/The-State-of-AI-2019-Divergence.pdf 8 years (Fig. 23), to an estimated $15bn in 2018 (CB Insights, MMC Ventures). Today’s leading technology companies – including Apple, Amazon, Facebook, Google, IBM, Microsoft and Salesforce – are also spending heavily on research and personnel to develop and deploy AI. Internal corporate investment on AI, among just the top 35 high tech and advanced manufacturing companies investing in AI, may be 2.0x to 4.5x greater than the capital invested by venture capital firms, private equity firms and other sources of external funding combined (McKinsey), further catalysing progress. have increased fifteen-fold in five years, to an estimated $15bn in 2018. (CB Insights, MMC Ventures) Source: CB Insights, MMC Ventures Fig 23. Venture capital investment in AI has increased 15-fold in five years 0 200 400 600 800 1000 1200 0 2 4 6 8 10 12 14 16 20132012 2014 2015 2016 2017 2018E Fig. X: Venture capital investment in AI has increased 15-fold in five years AIdeals Disclosed Funding (right axis) Number (left axis) AIdealinvestment($billion) High valuation
  • 9. 9 Fear Photo by samer daboul from Pexels
  • 10. It’s important to understand… 10 that these factors impact your AI project.
  • 11. What is AI? Strong AI vs. Weak AI AI is used as a generic term for a set of tools to cope with a certain set of problems. Machine Learning is a subset of this AI-toolset: „Programming by example“. Other subsets are knowledge representation, planning, reasoning. AI uses probabilistic logic instead of boolean logic. 11
  • 12. 12 „The brown quick fox jumps over the lazy dog“ https://www.mcohen.io/2017/machine-learning-explained-in-three-easy-steps/ brown quick
  • 13. Properties of AI problems Hard to code „by hand“ • Requires non-formalized knowledge (experiential knowledge) • Or even not yet existing knowledge Afflicted with uncertainty (or missing information) Changes rapidly (making it unreasonable to adopt software manually) 13
  • 14. Playing Chess, playing Go 14 f(s) f(s) f(s) f(s) f(s) f(s) f(s) f(s) 1 0 0 1
  • 15. 15Photo by Gratisography from Pexels digging deeper into „debugging“
  • 16. Recap: Properties of AI problems Hard to code „by hand“ • Requires non-formalized knowledge (experiential knowledge) • Or even not yet existing knowledge Afflicted with uncertainty (or missing information) Changes rapidly (making it unreasonable to adopt software manually) 16
  • 17. 17 These properties make verification hard „by design“ Recap: Properties of AI problems Hard to code „by hand“ • Requires non-formalized knowledge (experiential knowledge) • Or even not yet existing knowledge
  • 18. 18Photo by Pixabay from Pexels Problems with unknown truth For example: Medical classification problem What is the correct diagnosis? What is the correct therapy? You won’t know (maybe until your patient either recovers or dies?) Same is true for customer support systems. Is you customer satisfied? You’ll probably only know by loosing him/her as a customer.
  • 19. 19https://towardsdatascience.com/gender-bias-word-embeddings-76d9806a0e17, Photo by rawpixel.com from Pexels ML replicates bias in the data Example: Conceptual similarities from word embeddings With words typically collocated, you can ask you model for conceptual similarities: king - man + woman ⇾ queen Depending on your input corpus, your model will give you doctor - man + woman ⇾ nurse Do you really run a sexist, racist chatbot on your website?
  • 20. AI systems break fundamental patterns we developed as an industry • No (or very little) isolation. You need to verify and retrain the whole system. • Higher dimension of failure space • Time intense training cycles (instead immediate feedback cycles) • Non-stationary nature of ML systems 20http://ai.stanford.edu/~zayd/why-is-machine-learning-hard.html
  • 22. We need … 22 … new quality management approaches
  • 23. Quality management on input data Implement quality management for your input data. Visualize, use statistical metrics on the input data. Identify bias in the input data. Establish panel of judges both on input data labelling and on outcomes. Due to the non-stationary nature this is is an ongoing task, not a closed project. Use generated test data with known patterns, because otherwise you won’t know if you miss whole categories. 23
  • 24. Work with test sets To test for non-binary outcomes (i.e., results with confidence level), you need to handle test set as opposed to sets of single test outcomes: For example test for outcome confidence distribution. For non-stationary systems: establish test monitoring to accept regression. „Accept new model if result is at least 95% of last model.“ Do not only test outcomes, but implement inspection tools. E.g., in RoboCup Simulation map agent movement paths. 24
  • 25. Add a safety net Implement multi-layer security fallbacks for subsets of your problem (also to be used while testing), like „emergency break systems“. Test these. Use for testing: If you need your security fallback too often, your model may be bad. Just in research: Add explainability, local properties by black-box tests on the model to verify the „anchor rules" (https:// homes.cs.washington.edu/~marcotcr/aaai18.pdf) 25
  • 26. © data42 GmbH Speed / cost benefits of Machine Learning often only apply if you accept a non-zero failure rate Think twice before implementing high stakes AI applications (E.g., do NOT use AI to screen applications in HR) 26 26.02.18
  • 27. © data42 GmbH Derive new knowledge from AI … to solve new problems. 27 26.02.18
  • 28. Thank you very much, and be curious! xing.to/betz