This course covers processes and tools to minimize bias and unfairness in predictive models, particularly those using black-box algorithms. You will learn how to:
- Identify the types of unintended harm that can arise from AI models
- Explain why interpretability is key to avoiding harm
- Distinguish between intrinsically interpretable models and black box models
- Evaluate tradeoffs between model performance and interpretability
- Establish and implement a Responsible Data Science framework for your projects
- Evaluate predictor impact in black box models using interpretability methods
- Assess the performance of models with metrics to measure bias and unfairness
- Conduct an audit of a data science project from an ethical standpoint
Who Should Take This Course
Data Science architects and programmers, managers of data science projects and teams.
- Review of predictive modeling
- Why Responsible Data Science?
- Types of harms
- The black box problem
- Legal considerations, legal
- Why interpretability is an ethical issue
- Interpretability versus performance tradeoff
- Establishing baseline
- Intrinsically-interpretable algorithms
- Interpretability for black-box algorithms
- Global interpretability
- Local interpretability
- The RDS Framework
- Enhancing standard “best practices” from an ethical standpoint
- The 10 RDS questions to answer
- Metrics for assessing bias
- Assessment of results of applying local and global interpretability methods
- Auditing for neural nets (briefly)
You should be familiar with predictive modeling and able to work in R or Python. Either of the following courses is good preparation:
The Statistics.com courses have helped me a lot, pushing me to the limit and making me learn much more than I expected I could. The knowledge I gained I could immediately leverage in my job … then eventually led to landing a job in my dream company – Amazon.
This program has been a life and work game changer for me. Within 2 weeks of taking this class, I was able to produce far more than I ever had before.
The material covered in the Analytics for Data Science Certificate will be indispensable in my work. I can’t wait to take other courses. Great work!
I learned more in the past 6 weeks than I did taking a full semester of statistics in college, and 10 weeks of statistics in graduate school. Seriously.
This is the best online course I have ever taken. Very well prepared. Covers a lot of real-life problems. Good job, thank you very much!
The more courses I take at Statistics.com, the more appreciation I have for the smart approach, quality of instructors, assistants, admin and program. Well done!
This course greatly benefited me because I am interested in working in AI. It has given me solid foundational knowledge…After completing this last course, I feel I have gained valuable skills that will enhance my employability in Data Science, opening up diverse career opportunities.
Frequently Asked Questions
What is your satisfaction guarantee and how does it work?
Can I transfer or withdraw from a course?
Who are the instructors at Statistics.com?
Visit our knowledge base and learn more.
About 15 hours per week, at times of your choosing.
Homework consists primarily of practical exercises with R or Python.
The required text for this course is Responsible Data Science, Wiley, by Grant Fleming, Peter Bruce. Please order a copy of your course textbook prior to course start date.”
Python or R
Course Fee & Information
Courses may fill up at any time and registrations are processed in the order in which they are received. Your registration will be confirmed for the first available course date unless you specify otherwise.
Transfers and Withdrawals
We have flexible policies to transfer to another course or withdraw if necessary.
Contact us to get information on group rates.
Academic affiliation? In most courses you are eligible for a discount at checkout. Use promo code ACADEMIC where prompted during registration.
Invoice or Purchase Order
Add $50 service fee if you require a prior invoice, or if you need to submit a purchase order or voucher, pay by wire transfer or EFT, or refund and reprocess a prior payment.
Literacy, Accessibility, and Dyslexia
At Statistics.com, we aim to provide a learning environment suitable for everyone. To help you get the most out of your learning experience, we have researched and tested several assistance tools. For students with dyslexia, colorblindness, or reading difficulties, we recommend the following web browser add-ons and extensions:
- Color Enhancer (for colorblindness)
- HelperBird (for colorblindness, dyslexia, and reading difficulties)
- Mobile Dyslexic
- Color Vision Simulation (native accessibility feature)
- Other native accessibility features instructions