iiinova — where your digital self keeps evolving

Christoph Molnar @christophmolnar.bsky.social · 08/07/2025

Using feature importance to interpret your models? This paper might be of interest to you. Papers by @gunnark.bsky.social are always worth checking out.

15272

Christoph Molnar @christophmolnar.bsky.social · 08/04/2025

My stock portfolio is deep in the red, and tariffs by the Trump admin might be the cause. Could an LLM have been used to calculate them? It made me rethink how LLMs shape decisions, from big global-economy-wrecking ones to everyday decisions.

mindfulmodeler.substack.com

Who’s Really Making the Decisions?

LLMs, tariffs, and the silent takeover of decisions

3113

Christoph Molnar @christophmolnar.bsky.social · 01/04/2025

SHAP interpretations depend on background data — change the data, change the explanation. A critical but often overlooked issue in model interpretability. Read more:

mindfulmodeler.substack.com

SHAP Interpretations Depend on Background Data — Here’s Why

Or why height doesn't matter in the NBA

0121

Christoph Molnar @christophmolnar.bsky.social · 13/03/2025

The 3rd edition of Interpretable Machine Learning is out! 🎉 Major cleanup, better examples, and new chapters on Data & Models, Interpretability Goals, Ceteris Paribus, and LOFO Importance. The book remains free to read for everyone. But you can also buy ebook or paperback.

0163

Christoph Molnar @christophmolnar.bsky.social · 12/03/2025

Has anyone seen Counterfactual Explanations for machine learning models somewhere in the wild? They are often discussed in research papers, but I have yet to see them being used somewhere in an actual process or product.

1101

Christoph Molnar @christophmolnar.bsky.social · 12/03/2025

It's still hard to predict for me when it fails. For example, told it to simply check for placements of citations in a markdown file, which should be doable with a regex. And Claude failed. But a similar task worked out the other day.

120

Christoph Molnar @christophmolnar.bsky.social · 12/03/2025

Trying Claude Code for some tasks. Paradoxically, it's most expensive when it doesn't work because it fails, then tries a couple of times again, burning through tokens. So sometimes it's 20 cents for saving you 20 minutes of work. Other times it's $1 for wasting 10 minutes.

2763

Christoph Molnar @christophmolnar.bsky.social · 07/03/2025

Only waiting for the print proof, but if it looks good, I'll publish the third edition of Interpretable Machine Learning next week. As always, it was more work than anticipated—especially moving the entire book project from Bookdown to Quarto, which took a bit of effort.

1180

Christoph Molnar @christophmolnar.bsky.social · 25/02/2025

Can an office game outperform machine learning? My most recent post on Mindful Modeler dives into the wisdom of the crowds and prediction markets. Read the full story here:

buff.ly

Can an office game outperform machine learning?

Wisdom of the crowds, prediction markets, and more fun in the work place.

050

Christoph Molnar @christophmolnar.bsky.social · 04/02/2025

5/ It was stressful, but I don’t regret it. I learned a lot and definitely feel validated in my skills again. Full story & solution details: buff.ly/4gHZYHD

mindfulmodeler.substack.com

How to win an ML competition beyond predictive performance

A dive into the challenges and winning solution

0110

Christoph Molnar @christophmolnar.bsky.social · 04/02/2025

4/ Writing Supervised ML for Science at the same time was a huge plus—competition & book writing fed into each other (e.g., uncertainty quantification).

150

Christoph Molnar @christophmolnar.bsky.social · 04/02/2025

3/ One key insight: SHAP’s reference data matters! I used historical forecasts for interpretability. Also combined SHAP with ceteris paribus profiles for sensitivity analysis.

140

Christoph Molnar @christophmolnar.bsky.social · 04/02/2025

2/ My approach: ✅ XGBoost ensemble, quantile loss ✅ SHAP for explainability + custom waterfall plots + ceteris paribus plots ✅ Conformal prediction to fix interval coverage ✅ Auto-generated reports with Quarto

120

Christoph Molnar @christophmolnar.bsky.social · 04/02/2025

1/ Years ago, I went full-time into writing & cut ML practice. At some point, I felt like an impostor writing about ML but no longer practicing. This competition about water supply forecasting on DrivenData (500k prize pool) was a way back in.

120

Christoph Molnar @christophmolnar.bsky.social · 04/02/2025

A year ago, I took a risk & spent quite some time on a ML competition. It paid off—I won 4th place overall & 1st in explainability! Here's a summary of the journey, challenges, & key insights from my winning solution (water supply forecasting)

1221

Christoph Molnar @christophmolnar.bsky.social · 22/01/2025

deprecated was maybe the wrong word. It's no longer the default in the shap package. There are faster alternatives

040

Christoph Molnar @christophmolnar.bsky.social · 22/01/2025

The connection between SHAP and LIME is only when we represent features differently for LIME and use a different weight function. My take is that, while interesting, it can be misleading as SHAP and original LIME are very different, as you also say.

110

Christoph Molnar @christophmolnar.bsky.social · 22/01/2025

The original SHAP paper has been cited over 30k times. The paper showed that attribution methods, like LIME and LRP, compute Shapley values (with some adaptations). The paper also introduces estimation methods for Shapley values, like KernelSHAP, which today is deprecated.

3262

Christoph Molnar @christophmolnar.bsky.social · 21/01/2025

To this day, the Interpretable Machine Learning book is still my most impactful project. But as time went on, I dreaded working on it. Fortunately, I found the motivation again and I'm working on the 3rd edition. 😁 Read more here:

buff.ly

Why I almost stopped working on Interpretable Machine Learning

7 years ago I started writing the book Interpretable Machine Learning.

1213

Christoph Molnar @christophmolnar.bsky.social · 14/01/2025

How I sometimes feel working on "traditional" machine learning topics instead of generative AI stuff 😂

the "They don't know meme". Party. Awkward guy in the corner saying They don't know I work on the 3rd edition of Interpretable Machine Learning". Rest is having fun, titled "Everyone else talking about generative AI"

1415

Christoph Molnar @christophmolnar.bsky.social · 09/01/2025

Okay, that makes a lot of sense. Thanks!

020

Christoph Molnar @christophmolnar.bsky.social · 09/01/2025

I love that visualization. At the same time confused about why souping works.

120

Christoph Molnar @christophmolnar.bsky.social · 09/01/2025

While the results on ARC are impressive, the ARC problems are abstract games in little grid worlds. And it remains to be seen whether this translates into "general" intelligence.

110

Christoph Molnar @christophmolnar.bsky.social · 17/12/2024

Yeah, I'm mixing two very different scenarios here. The socioeconomic ones are even tougher, so predicting the first place is extremely speculative.

010

Christoph Molnar @christophmolnar.bsky.social · 17/12/2024

It's quite ironic how people who built the best prediction models are such bad predictors themselves. They throw all their knowledge about how to make good predictions overboard and just claim things like: AI will replace radiologists in a few years or when they expect AGI.

1211

Christoph Molnar @christophmolnar.bsky.social · 17/12/2024

As a consequence, there is disappointment.

170

Christoph Molnar @christophmolnar.bsky.social · 17/12/2024

The problem with all these AI demos (especially for image and video generation): They are the most impressive, cherry-picked examples. That includes cherry-picking prompts and themes that produced better results. But as a user, you want good results for every prompt/theme relevant to your use case

3211

Christoph Molnar @christophmolnar.bsky.social · 14/12/2024

Because I have multiple books already in Quarto, and it makes it easier to maintain them when they share the same software. Also I find Quarto easier to handle

010

Christoph Molnar @christophmolnar.bsky.social · 13/12/2024

Wait till he checks the google trend for XAI.

040

Christoph Molnar @christophmolnar.bsky.social · 13/12/2024

Alternatively, you only read the "not all you need papers"

020

Christoph Molnar @christophmolnar.bsky.social · 13/12/2024

Looking for a Christmas gift for a stubborn Bayesian or an over-hyped AI enthusiast? Modeling Mindsets is a short read to broaden your perspective on data modeling. christophmolnar.com/books/modeli... *Hat not included.

Cover of Modeling Mindsets book. The octopus on the cover has a Christmas hat on. Still looks a bit creepy though.

3112

Christoph Molnar @christophmolnar.bsky.social · 13/12/2024

True for almost eradicated diseases. For other diseases, the upside is also personal: For example, in my area ticks can give you Tick-borne encephalitis, but since I have the vaccine, I feel much saver. But maybe for some this upside is a bit abstract, since you might never get the disease anyways.

110

Christoph Molnar @christophmolnar.bsky.social · 13/12/2024

My personal rules for AI-assisted writing: • Use AI only for small and specific stuff, like grammar fixes or making suggestions for factual corrections. • Never let an LLM change voice and tone. • I review any changes made by AI.

2141

Christoph Molnar @christophmolnar.bsky.social · 13/12/2024

Actually sitting here with a bit of knee pain while writing this 😂

100

Christoph Molnar @christophmolnar.bsky.social · 13/12/2024

What a sad timeline, where vaccines — one of medicine's clearest wins with all upside and minimal downside — have become targets. Can't we have like an anti-knee arthroscopy movement or whatever instead?

3140

Christoph Molnar @christophmolnar.bsky.social · 04/12/2024

Citing a non-deterministic, "hallucinating", and non-reproducible LLM output is wild. While the norms and best practices are evolving, citing them seems the wrong way. (even wilder when some people add "ChatGPT" as their co-authors)

4223

Christoph Molnar @christophmolnar.bsky.social · 03/12/2024

What are shapley interactions and why should you care about them? This is a guest post by Julia, Max, Fabian and Hubert on my newsletter Mindful Modeler. I also learned a lot from this post and definitely recommend checking out the shapiq package. mindfulmodeler.substack.com/p/what-are-s...

mindfulmodeler.substack.com

What Are Shapley Interactions, and Why Should You Care?

A guest post by Julia, Max, Fabian and Hubert.

1362

Christoph Molnar @christophmolnar.bsky.social · 03/12/2024

Why an entire weekend? A couple of TikTok videos should do 🤡

040

Christoph Molnar @christophmolnar.bsky.social · 29/11/2024

Is anyone aware of a completely AI-generated book that people actually read? Excluding books that are dedicated "AI experiments" and where the book is more about the experiment. Also excluding AI-assisted books where generative AI played a minor role

190