Home » Technical Topics » Data Science

No Causation without representation!

My free book Bayesuvius now has 39 chapters. I had been postponing writing the chapters on Pearl causality until now, because I consider them to be the most important chapters in the whole book, and I wanted to nail them, to the best of my limited abilities. Well, I finally bit the bullet and wrote them. Please check them out, and send me feedback. I would especially like your opinion on  the chapters entitled

  • D-Separation
  • Do-Calculus
  • Counterfactual Reasoning

Here is a picture from the Counterfactual Reasoning chapter to pique your interest:


Please keep in mind that this is the **first** released version of these chapters. I intend to improve and expand them in the future.

I truly believe that causal AI is an essential, fundamental part of AI, and that current AI will never progress beyond dumb curve fitting unless it embraces causal AI fully. That is why I am so delighted to see the recent meteoric rise of  the mega startup Insitro, which is applying (causal) AI to finding and testing new drugs. Causal reasoning is ubiquitous in human thinking. Even cats do it when they investigate a water faucet. And yet current AI doesn’t use it! I end this post with my causal chain of the day:

No causal AI without Bayesian Networks representation. No Bayesuvius without causal AI.

bnet representation ——>causal AI——>Bayesuvius