Andrew Laack

research

My main research interest is classical machine learning. In particular, oblique decision trees, ensemble methods, and interpretability.

publications

CART-ELC: Oblique decision tree induction via exhaustive search

https://doi.org/10.48550/arxiv.2505.05402

code

I like building software and implementing machine learning research. My code is self hosted. I have some code available on my Github, but I am not a fan of the platform becoming the de facto standard for code distribution given its proprietary nature.

https://git.laack.co/

https://github.com/andrewlaack

studies

I graduated from the University in Wisconsin-Superior with a BS in CS in the Spring of 2025. I am also starting a Master's degree program at Georgia Tech in the Fall of 2025. I have taken a fair number of computer science courses and math courses, but I have found it much more rewarding to self-study. Below is a non-comprehensive list of books I have read and recommend to those interested in machine learning.

ml book recommendations (in order)

math book recommendations (unordered)

These were the most impactful math and machine learning books that I read early on. For those interested in reading them, I would recommend first reading Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow to get a good understanding of what machine learning is all about and how to use modern machine learning tools. I would then start reading the math books to gain a strong understanding of linear algebra, probability, and calculus, prior to returning to the more theory based machine learning books. Once all of these have been read and well understood, it becomes evident what is of interest to the reader and thus their paths may diverge, but I think these recommendations are non-controversial.

blog

On my blog I discuss my views about software, freedom, and other topics of interest.

gemini://blog.laack.co/

contact

My email is andrew@laack.co.