<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Blaz Stojanovic</title><description>Notes on complex systems, reasoning, and the art of pre-training.</description><link>https://blazstojanovic.github.io/bstojanovic/</link><language>en-us</language><item><title>The unreasonable effectiveness of decision trees for tabular data</title><link>https://blazstojanovic.github.io/bstojanovic/blog/tabular-incumbents/</link><guid isPermaLink="true">https://blazstojanovic.github.io/bstojanovic/blog/tabular-incumbents/</guid><description>Gradient-boosted decision trees dominated the 2010s on tabular data, and are still extremely prevalent today. We examine the reasons behind their success, and continued relevance in the 2020s.</description><pubDate>Mon, 04 May 2026 00:00:00 GMT</pubDate><category>tabular</category><category>gbdt</category><category>xgboost</category><category>lightgbm</category><category>catboost</category></item></channel></rss>