2025-11-05 –, ML+analytics
A couple of months ago I got married! Very exciting, and a great excuse to complicate things by writing code.
One of the ways Python was involved is in creating abstract backgrounds for copy such as the invitation, menus and so on, using pictures we took in our travels. In the talk I'll walk through the journey of finding the technique to artistically recreate images with decision trees.
Everybody loves algorithmic art, and what could be better than data-driven algorithmic art?
When my wife and I got married, we wanted to add our touch to as many aspects of the wedding as we could (what, control freaks? Us? No way).
We thought it would be cool to design the graphic elements - invitation, menus, etc. - ourselves, and I really wanted to use this opportunity to play with algorithmic art based on our pictures.
After a lot of trial and error, I landed on a technique which gave nice results, using decision trees / random forests, and we ended up using it!
In this talk I'll demonstrate the idea, walk through the process of arriving at it, and provide general thoughts for starting with data-driven art.
No previous knowledge expected
Hi! I'm Daniel, a machine learning research engineer from Israel.
I started my career as a programmer, and had the opportunity to take on a range of roles: FS developer, team lead, and course instructor.
I then got into the world of data science / ML in Nutrino - a nutrition startup that was later acquired by Medtronic.
After that I ventured out on my own as a freelance machine learning practitioner; now I work with several startups on ML / vision problems. I also work with an Israeli non-profit, helping to create and improve tech-ed programs for youth.
Bachelor's degree in computer science from the Open University, Master's degree in machine learning & data science from Reichman University.
Partial list of things I'm excited about:
- Algorithms and optimizations
- Translating experts’ domain knowledge and intuition into concrete methods and metrics
- Designing ad-hoc models that make the most of little data (mostly because that involves designing custom metrics and coming up with creative ways of optimizing them)
- Finding patterns in behavior
- Creating interactive visualizations to explore complex data and interactions
Also, I have a blog!