Blog

Long-form research notes from machine learning work: search, evaluation, model behavior, and interpretability.

Benchmarks 5 min read

Which Machine Learning Classifiers are Best for Small Datasets?

An empirical pass across 108 benchmark datasets to see what actually holds up when you only have 100–1,000 rows.

Read the featured post

Why two models with the same held-out score can still behave very differently in production.

Lessons from using three years of search logs to improve Semantic Scholar’s ranking system.

Why variance changes the way feature importance shows up, and how to explain that clearly.