2023-07-30

2023-07-30 Sunday - Book Review: Causal Inference and Discovery with Python

 

[image source: Amazon.com]

Causal Inference and Discovery with Python: Unlock the secrets of modern causal machine learning with DoWhy, EconML, PyTorch and more
https://www.amazon.com/Causal-Inference-Discovery-Python-learning/dp/1804612987  

[My Amazon posted review: link]

My Review Title

A Fascinating Book on Causal Machine Learning -  and the best Packt published book I’ve ever read

My Review Summary:

Why should you want to read this book?

Understand the landscape of causality modeling – and a broad survey of the many available tools – as well as the historical background - and fairly recent trends - in causality modeling research.

To learn the reason to use causal modeling rather than traditional machine learning.

Understand the libraries and algorithms available for causal modeling.

Understand the use cases (and limitations) of different causal modeling strategies and capabilities.

Understand the importance (and challenges) of obtaining Expert Knowledge for causal modeling.

Understand the practical aspects of applying and implementing causality modeling to business problems.

This book will save you months of effort trying to research and assemble the equivalent information. The book offers an optimal and efficient path to learning the information.

What I particularly liked:

  • Breadth and depth of the treatment of the material.
  • Inclusion of interesting examples – written in Python.
  • Use of Jupyter notebooks
  • The excellent quality of the images for source code and graphics.
  • Inclusion of the link to the companion github repository for the book
  • QR code to download a PDF version of the book
  • Inclusion of a References section at the end of chapters, with suggested additional reading. This *really* enriches a book for me – and indicates that the author has given deep thought to how to expand and elevate the reader’s understanding of the material.
  • The treatment and depth of the material covered – is exceptionally well-done.
  • The writing is ACCESSIBLE. The author invests sufficient effort to bring the reader along – and build their knowledge with successive layering of concepts.
  • The material covered is fascinating - well-researched - and every single chapter is filled with rich content - as well as extensive citations in the References section, at the end of each chapter, with *numerous* high quality suggestions for additional reading.
  • The inclusion of excellent coding examples.
  • The writing is exceptionally well-organized, CRISP (a word I reserve for only the very best writing) – and weaves concepts, theory, and practical hands-on coding exercises – into a seamless narrative.
  • The writing is fresh – and lively – filled with insights and examples that are meaningful to any reader interested in Causality Inference/Discovery/Modeling, Machine Learning, and Deep Learning.
  • Near the end of the book, I came to a new appreciation for the care with which the author has taken to prepare and organize the information – in successive elegant layers – to enrich the reader’s learning experience. This is another confirmation of the exceptional quality of the writing – and the author’s depth of preparation to write this book.

Taken as a whole (the combination of the writer’s content – and the cornucopia of suggested follow-up references) – this book easily wins, hands-down, as the best book I’ve ever read, published by Packt.

From Amazon:
https://www.amazon.com/Causal-Inference-Discovery-Python-learning/dp/1804612987  

Demystify causal inference and casual discovery by uncovering causal principles and merging them with powerful machine learning algorithms for observational and experimental data

Causal methods present unique challenges compared to traditional machine learning and statistics. Learning causality can be challenging, but it offers distinct advantages that elude a purely statistical mindset. Causal Inference and Discovery in Python helps you unlock the potential of causality.

You'll start with basic motivations behind causal thinking and a comprehensive introduction to Pearlian causal concepts, such as structural causal models, interventions, counterfactuals, and more. Each concept is accompanied by a theoretical explanation and a set of practical exercises with Python code.

Next, you'll dive into the world of causal effect estimation, consistently progressing towards modern machine learning methods. Step-by-step, you'll discover Python causal ecosystem and harness the power of cutting-edge algorithms. You'll further explore the mechanics of how “causes leave traces” and compare the main families of causal discovery algorithms.

The final chapter gives you a broad outlook into the future of causal AI where we examine challenges and opportunities and provide you with a comprehensive list of resources to learn more.

What you will learn:

  • Master the fundamental concepts of causal inference
  • Decipher the mysteries of structural causal models
  • Unleash the power of the 4-step causal inference process in Python
  • Explore advanced uplift modeling techniques
  • Unlock the secrets of modern causal discovery using Python
  • Use causal inference for social impact and community benefit


From Packt:
https://www.packtpub.com/product/causal-inference-and-discovery-in-python/9781804612989

"Causal methods present unique challenges compared to traditional machine learning and statistics. Learning causality can be challenging, but it offers distinct advantages that elude a purely statistical mindset. Causal Inference and Discovery in Python helps you unlock the potential of causality. You’ll start with basic motivations behind causal thinking and a comprehensive introduction to Pearlian causal concepts, such as structural causal models, interventions, counterfactuals, and more. Each concept is accompanied by a theoretical explanation and a set of practical exercises with Python code. Next, you’ll dive into the world of causal effect estimation, consistently progressing towards modern machine learning methods. Step-by-step, you’ll discover Python causal ecosystem and harness the power of cutting-edge algorithms. You’ll further explore the mechanics of how “causes leave traces” and compare the main families of causal discovery algorithms. The final chapter gives you a broad outlook into the future of causal AI where we examine challenges and opportunities and provide you with a comprehensive list of resources to learn more.

 

Table of Contents

1.    Causality – Hey, We Have Machine Learning, So Why Even Bother?
2.    Judea Pearl and the Ladder of Causation
3.    Regression, Observations, and Interventions
4.    Graphical Models
5.    Forks, Chains, and Immoralities
6.    Nodes, Edges, and Statistical (In)dependence
7.    The Four-Step Process of Causal Inference
8.    Causal Models – Assumptions and Challenges
9.    Causal Inference and Machine Learning – from Matching to Meta-Learners
10.    Causal Inference and Machine Learning – Advanced Estimators, Experiments, Evaluations, and More
11.    Causal Inference and Machine Learning – Deep Learning, NLP, and Beyond
12.    Can I Have a Causal Graph, Please?
13.    Causal Discovery and Machine Learning – from Assumptions to Applications
14.    Causal Discovery and Machine Learning – Advanced Deep Learning and Beyond
15.    Epilogue


2023-07-30 Sunday - Are Backlogs Useless?

This was my reply to Wolfrum Müller's LinkedIn post (re: "Backlogs as the Blind Spot of Agile at Scale").

 

Are backlogs useless, or evil?


No. It is how they are misused to de-prioritize higher value innovative work that begets their abuse.

Discarding backlogs is like throwing away a useful tool just because it isn't a hammer.

At the poker table of business-driven priorities - technical debt will rarely beat whatever the business says is the next-most-important-thing. And often, for some period of time, technical debt often can only be characterized as something that *may* threaten the business at some point in the future...

And yet, technical debt needs to be tracked and planned - or some variation of it will bite you in the ass, destroy the ability of your business to compete, create conditions for security breaches, hinder/prevent your ability to innovate/execute, create massive obstacles to efficiently running/maintaining/modifying systems, etc.

Hence, backlog.

For in them - there is important (and quite often, critical) information. Just because the business doesn't see value in something - doesn't mean it is necessarily trash to be discarded and thrown out.

[image credit: PublicDomainPictures on pixabay.com]

 

 

Backlogs are not just necessary at the team level, but also at the enterprise level:

Teams are often too narrowly focused, too narrowly scoped, and too limited in their charter/funding - and thus, there is a need for an enterprise-view backlog that crosses domains, business units, organizational silos, etc.

For example:

  • Platform / Shared Services
  • Information Security / Security Architecture
  • Information & Data Architectures
  • Integration Architecture
  • Business Architecture
  • Infrastructure/Cloud Architecture
  • Enterprise Vision / Strategy / Roadmap / Runways



 

2023-07-29

2023-07-30 Saturday - Consistent Intentional Variation and Experimentation

 

[image credit: Schäferle on pixabay.com]
 

Today's meditation:

In the worlds of music and sports - beginners are taught basic drills.

For most students, the majority of those drills will be repeated- week after week, month after month, year after year. This will usually produce deep connections - and elevate the student's ability to respond naturally.

Technically, these students may be well trained - but they will rarely be exceptional. Their abilities rarely evolve beyond rote repetition.

There is a parallel to this, in how most people perform their jobs - day after day, month after month, year after year.

However, some teachers/coaches will introduce slight variations in the drills - rarely repeating the exact same drill. This subtle difference in approach invariably produces students that progress faster, have more fluid and natural responses - and are more adept when confronted with new and novel situations.

Seeking consistent intentional variations in what we do, in how we do what we do - can help elevate our skills much faster, and more broadly - than we might imagine.

Variation and experimentation is the secret to mastery of any skill, or area of knowledge.

Embracing variation is the harder path - it forces us to break out of our comfortable rhythms. It is to embrace a continual cycle of renewal - and seeking our original Beginner's Mind.

2023-07-23

2023-07-23 Sunday - EA Modeling Tools: Readiness Assessment & Selection Guidance

 

book cover, (c) 2023 Kelvin D. Meeks

Has your organization tried - and failed - to establish an EA Modeling Tool?

Has the failure of some other organization - to successfully establish an EA Modeling Tool - prevented you from being able to convince your leadership team to adopt one?

Do you feel overwhelmed by the internal arguments against an EA Modeling Tool - and wish that you had the time to craft a comprehensive set of arguments for the WHY?

Do you wish that you had someone on your team that has "done it before"? 

Someone like a "mountain guide" to help you and your team "plan your route" - and avoid the problems that could derail your efforts?

That's why I'm writing this book.

 

I was originally intending to just write a blog post - but as I drafted the outline of the blog post - I realized that I would need to actually write a book to provide sufficient treatment of the material. 

Today, just the outline - is 26 pages...and will continue to evolve. 

I plan to self-publish the book before October 1st, 2023 - so check back on this page for more details on how you can purchase the book.

If you would like to be notified when the book is available for purchase, you can drop me an email (kmeeks@intltechventures.com)

2023-08-05 Saturday - update

  • Rough layout is now 90 pages, 18 chapters + References & Bibliography
  • While pondering what price to set for the book - I turned to the wisdom of Douglas Adams ("the answer to the ultimate question of life, the universe, and everything") - and will sell it for $42 (also the amount it would cost for someone to treat me to a mocha for a week at my favorite local coffee shop...which seems like a fair trade for the decades of knowledge and experience I'm sharing in the book).

2023-07-17

2023-07-17 Monday - An Ode to SharePoint

[image credit: fietzfotos on pixabay.com]

 

 Today's meditation:
(an ode to SharePoint...as a 4-part haiku 🤣)

SharePoint. Bane to All.
Like a Plague, You Torment Us.
May Death Find You, Soon.

Hinder & Hamper.
Not a Wiki - Poor Fake Tool.
Blight Upon IT.

Brittle By Design,
Authoring Pages - Woeful.
Your Broken Links Rot.

In a Parking Lot,
Before My Car Might You Be.
Ramming Speed, I Say.

 

There are some tools I dislike, some that I love...and some tools I absolutely despise. SharePoint, look at me when I'm talking to you! 🤣 

Now class, repeat after me: "SharePoint is NOT an EA Repository"


2023-07-02

2023-07-02 Sunday - On The Importance of Taking Notes

 

[image credit: Engin_Akyurt on pixabay.com]

This blog post is my follow-up comment/reply to Werner's recent article

Today, I noted Werner's post on LinkedIn....

2023-06-30 Werner Vogels: A few words on taking notes
https://www.linkedin.com/posts/wernervogels_a-few-words-on-taking-notes-activity-7080158996879310848-eAGc
https://www.allthingsdistributed.com/2023/06/a-few-words-on-taking-notes.html
"As we are about to start the planning meetings for 2024 at AWS, I’ve been thinking a lot about how I take notes."

Here are some of the article citations I've collected over the years...

2021-03-19: Paper Notebooks vs. Mobile Devices: Brain Activation Differences During Memory Retrieval
Front. Behav. Neurosci., 19 March 2021
Sec. Learning and Memory
Volume 15 - 2021 | https://doi.org/10.3389/fnbeh.2021.634158
https://www.frontiersin.org/articles/10.3389/fnbeh.2021.634158/full
A recent study by Kuniyoshi Sakai, titled Paper Notebooks vs. Mobile Devices: Brain Activation Differences During Memory Retrieval, actually showed higher retention and recall for subjects that used pen and paper versus a keyboard, or tablet and stylus.


2017-07-07 HBR: The More Senior Your Job Title, the More You Need to Keep a Journal, by Dan Ciampa
https://hbr.org/2017/07/the-more-senior-your-job-title-the-more-you-need-to-keep-a-journal


2015-02-19 Richard Branson (LinkedIn post): The Importance Of Taking Notes
https://www.linkedin.com/pulse/importance-taking-notes-richard-branson/


2014-07-04 Psychology Today: Step Away From The Keyboard: How Our Hands Affect Our Brains
https://www.psychologytoday.com/intl/blog/thinking-about-kids/201407/step-away-the-keyboard-how-our-hands-affect-our-brains
Nancy Darling, Ph.D., professor of psychology at Oberlin College.


2014-06-03 Scientific American: A Learning Secret: Don't Take Notes with a Laptop
https://www.scientificamerican.com/article/a-learning-secret-don-t-take-notes-with-a-laptop/
"New research by Pam Mueller and Daniel Oppenheimer (Princeton, UCLA) demonstrates that students who write out their notes on paper actually learn more."


2014-04-23 Psychological Science (Volume 25, Issue 6): The Pen Is Mightier Than the Keyboard: Advantages of Longhand Over Laptop Note Taking
https://journals.sagepub.com/doi/10.1177/0956797614524581
https://doi.org/10.1177/0956797614524581


2014-09-12 The Benefits Of Writing With Good Old Fashioned Pen And Paper
https://www.huffpost.com/entry/writing-on-paper_n_5797506
learning and remembering course material, the pen is mightier than the keyboard,” according to Indiana University’s School of Medicine.
https://www.academia.edu/6273095/The_Pen_Is_Mightier_Than_The_Keyboard_Advantages_of_Longhand_Over_Laptop_Note_Taking


2011-01-24 Better learning through handwriting, The University of Stavanger
https://www.sciencedaily.com/releases/2011/01/110119095458.htm
Written by Trond Egil Toft; translation by Astri Sivertsen


2010-04-01 Digitizing Literacy: Reflections on the Haptics of Writing
https://www.intechopen.com/chapters/9927


2007-09-04 Workshop "Supporting Human Memory with Interactive Systems"
HCI Conference, Lancaster, UK
Does Taking Notes Help You Remember Better? Exploring How Note Taking Relates to Memory
https://diuf.unifr.ch/people/lalanned/MeMos07/files/kalnikaite.pdf


https://en.wikipedia.org/wiki/Kinesthetic_learning

https://en.wikipedia.org/wiki/Encoding_(memory)

https://en.wikipedia.org/wiki/Modality_effect

https://en.wikipedia.org/wiki/Transfer-appropriate_processing
 

2023-07-02 Sunday - Don't Play Like a Kindergartener

[image credit: _Alicja_ on pixabay.com]


 I saw someone's post on LinkedIn this morning:

A few highlights from the post:

  • In a famous experiment, Kindergarteners outperformed CEOs, Lawyers, and MBA
  • THE EXPERIMENT: Each group was asked to build the tallest tower they could
    • 26” - Kindergarteners
    • 22” - CEOs
    • 15” - Lawyers
    • 10” - MBA students
  • MBA Students in Action: Start thinking and talking strategically.
  • Kindergartners in Action: No strategy, question or discussion phase.
  • The Kindergarteners don’t waste energy. They just experiment.

  • Play like a kindergartener
     

My response:

 I'll take a counterpoint view for a moment and propose some dimensions to consider that may help illustrate the problems with adopting such simplistic views of a toy experiment that didn't deeply consider the 2nd and 3rd order effects of its practical implications:
  • Complex change has dependencies - within teams, across organizational business units, and across organizational boundaries.
  • Achieving complex change requires planning, communicating, coordination, alignment, and selling - as well as developing consensus to obtain funding.
  • Real organizations, in real industries, with real consequences - must adhere to a myriad tangle of regulations, standards, and compliance requirements. Real organizations have limitations (resources, time, budgets) - they don't have the luxury for unlimited experiments and retries. To ensure that a real organization doesn't waste its resources, operates efficiently, and meets its regulatory and compliance obligations - requires governance.
  • Without governance, every individual or team - may choose a different approach, resulting in duplication of effort and resources - resulting in massive waste and rework (again, and again, and again).

Be thoughtful - don't be mindless drones.

 

2016-06-21 HBRStrategic Plans Are Less Important than Strategic Planning
by Graham Kenny

As Kenny noted, Winston Churchill and Dwight D. Eisenhower had similar views. Churchill said, “Plans are of little importance, but planning is essential,” while Eisenhower said, “Plans are worthless, but planning is everything.”

Copyright

© 2001-2021 International Technology Ventures, Inc., All Rights Reserved.