Algorithmic auditing 1.0
0 likes165 views
The document discusses algorithmic auditing within the context of AI and machine learning in finance, emphasizing the importance of validating algorithms, data, and processes for effective implementation. It outlines five key aspects to consider during auditing, including use cases, data quality, model evaluation, operational environment, and auditing processes. Sri Krishnamurthy, the presenter, highlights the relevance of algorithmic auditing in addressing biases, ensuring fairness, and enhancing interpretability in financial applications.
Algorithmic auditing 1.0
- 1. Pragmatic Algorithmic Auditing 1.0 2020 Copyright QuantUniversity LLC. Presented By: Sri Krishnamurthy, CFA, CAP sri@quantuniversity.com www.quantuniversity.com Data Club Northeastern University
- 2. 2 Speaker bio • Advisory and Consultancy for Financial Analytics • Prior Experience at MathWorks, Citigroup and Endeca and 25+ financial services and energy customers. • Columnist for the Wilmott Magazine • Author of forthcoming book “Pragmatic AI and ML in Finance” • Teaches AI/ML and Fintech Related topics in the MS and MBA programs at Northeastern University, Boston • Reviewer: Journal of Asset Management Sri Krishnamurthy Founder and CEO QuantUniversity
- 3. 3 QuantUniversity • Boston-based Data Science, Quant Finance and Machine Learning training and consulting advisory • Trained more than 1000 students in Quantitative methods, Data Science and Big Data Technologies using MATLAB, Python and R • Building a platform for AI and Machine Learning Experimentation
- 4. 1. Key trends in AI, Machine Learning & Fintech 2. Algorithmic Auditing – Introduction 3. 5 things to note when auditing an algorithm 1. Use case 2. Data 3. Model 4. Environment 5. Process Agenda
- 5. AI and Machine Learning in Finance
- 6. 6 The 4th Industrial revolution is Here! Source: Christoph Roser at AllAboutLean.com As per Wikipedia*, “The 4th Industrial Revolution ….. marked by emerging technology breakthroughs in a number of fields, including robotics, artificial intelligence, nanotechnology, quantum computing, biotechnology, the Internet of Things, the Industrial Internet of Things (IIoT), decentralized consensus, fifth-generation wireless technologies (5G), additive manufacturing/3D printing and fully autonomous vehicles.” * https://en.wikipedia.org/wiki/Fourth_Industrial_Revolution
- 7. 7 Scientists are disrupting the way we live! Source: https://www.ladn.eu/tech-a-suivre/mobilite-2030-vehicules-volants-open-data/
- 8. 8 Interest in Machine learning continues to grow https://www.wipo.int/edocs/pubdocs/en/wipo_pub_1055.pdf
- 9. 9 MACHINE LEARNING AND AI IS REVOLUTIONIZING FINANCE
- 10. 10
- 11. 11 Algorithm Audits in the news
- 13. 13 • Machine learning is the scientific study of algorithms and statistical models that computer systems use to effectively perform a specific task without using explicit instructions, relying on patterns and inference instead1 • Artificial intelligence is intelligence demonstrated by machines, in contrast to the natural intelligence displayed by humans and animals1 Defining Machine Learning and AI 13 1. https://en.wikipedia.org/wiki/Machine_learning 2. Figure Source: http://www.fsb.org/wp-content/uploads/P011117.pdf
- 14. 14 Machine Learning & AI in finance: A paradigm shift 14 Stochastic Models Factor Models Optimization Risk Factors P/Q Quants Derivative pricing Trading Strategies Simulations Distribution fitting Quant Real-time analytics Predictive analytics Machine Learning RPA NLP Deep Learning Computer Vision Graph Analytics Chatbots Sentiment Analysis Alternative Data Data Scientist
- 15. 15 The Virtuous Circle of Machine Learning and AI 15 Smart Algorithms Hardware Data
- 16. 16 The rise of Big Data and Data Science 16 Image Source: http://www.ibmbigdatahub.com/sites/default/files/infographic_file/4-Vs-of-big-data.jpg
- 17. 17 Smart Algorithms 17 Distributing Computing Frameworks Deep Learning Frameworks 1. Our labeled datasets were thousands of times too small. 2. Our computers were millions of times too slow. 3. We initialized the weights in a stupid way. 4. We used the wrong type of non-linearity. - Geoff Hinton “Capital One was able to determine fraudulent credit card applications in 100 milliseconds”* * http://go.databricks.com/hubfs/pdfs/Databricks-for-FinTech-170306.pdf
- 18. 18 Hardware Speed up calculations with 1000s of processors Scale computations with infinite compute power
- 19. Machine Learning Workflow Data Scraping/ Ingestion Data Exploration Data Cleansing and Processing Feature Engineering Model Evaluation & Tuning Model Selection Model Deployment/ Inference Supervised Unsupervised Modeling Data Engineer, Dev Ops Engineer Data Scientist/Quants Software/Web Engineer • AutoML • Model Validation • Interpretability Robotic Process Automation (RPA) (Microservices, Pipelines ) • SW: Web/ Rest API • HW: GPU, Cloud • Monitoring • Regression • KNN • Decision Trees • Naive Bayes • Neural Networks • Ensembles • Clustering • PCA • Autoencoder • RMS • MAPS • MAE • Confusion Matrix • Precision/Recall • ROC • Hyper-parameter tuning • Parameter Grids Risk Management/ Compliance(All stages) Analysts & Decision Makers
- 20. 20
- 21. 21 • Algorithmic auditing is a structured process conducted internally or by a qualified independent third party that involves: ▫ Verifying and/or validating the working of the algorithm along with the data, model, environment, process contextual to the use-cases in which the algorithm is supposed to be used. ▫ Identification of issues that are clearly articulated and scoped for the algorithm. – Examples include : bias, discrimination, explainability, interpretability etc. ▫ Documentation of the understanding of the algorithm’s behavior, uses as observed and evaluated by a qualified individual. ▫ Recommendations for mitigation, control and elimination of noted risks. Algorithmic Auditing
- 23. 23
- 24. 24 Questions to ask: • Do we really need this algorithm? • How will this algorithm be used? • Who/What will it affect? 1. Use cases are important
- 25. 25 Things to think about: • How much data do we have? • How will this affect the model? • Do we have enough data? • Are their privacy concerns? 2. Don’t forget the data
- 26. 26 All scenarios haven’t played out • Stress scenarios • What-if scenarios Challenges with real datasets Figure ref: http://www.actuaries.org/CTTEES_SOLV/Documents/StressTestingPaper.pdf
- 27. 27
- 28. 28 Questions to ask • Blackbox/Whitebox • Does the model work? • How do we handle imbalanced classes? • Is it fair/biased? • Can you explain the model? 3. Model Audit
- 29. 29 4. Environment Audit : Where will the model run?
- 30. 5.0 Process audit Data Scraping/ Ingestion Data Exploration Data Cleansing and Processing Feature Engineering Model Evaluation & Tuning Model Selection Model Deployment/ Inference Supervised Unsupervised Modeling Data Engineer, Dev Ops Engineer Data Scientist/Quants Software/Web Engineer • AutoML • Model Validation • Interpretability Robotic Process Automation (RPA) (Microservices, Pipelines ) • SW: Web/ Rest API • HW: GPU, Cloud • Monitoring • Regression • KNN • Decision Trees • Naive Bayes • Neural Networks • Ensembles • Clustering • PCA • Autoencoder • RMS • MAPS • MAE • Confusion Matrix • Precision/Recall • ROC • Hyper-parameter tuning • Parameter Grids Risk Management/ Compliance(All stages) Analysts & Decision Makers
- 31. Register at https://algoauditing.splashthat.com/ Classes start April 1st 2021 31
- 32. Thank you! Sri Krishnamurthy, CFA, CAP Founder and CEO QuantUniversity LLC. srikrishnamurthy www.qu.academy Contact Information, data and drawings embodied in this presentation are strictly a property of QuantUniversity LLC. and shall not be distributed or used in any other publication without the prior written consent of QuantUniversity LLC. 32