Interestingness First Classifiers - Data Processing Club

Share This PostMost existing machine learning models aim to maximize predictive accuracy, but in this article, I will introduce classifiers that prioritize interestingness. What Does It Mean to Prioritize Interestingness? For example, let us consider the task of classifying whether a user is an adult based on their profile.If the profile contains an age feature, […]

Share This Post Most existing machine learning models aim to maximize predictive accuracy, but in this article, I will introduce classifiers that prioritize interestingness . What Does It Mean to Prioritize Interestingness? For example, let us consider the task of classifying whether a user is an adult based on their profile. If the profile contains an age feature, then the rule “age ≥ 18” would achieve 100% accuracy. On social media, there is always a possibility that users write down a false age, but in most cases people report their age more or less correctly, so this rule would likely achi

saved by

• Jack Hogan

related reading

• Toy Models of Superpositiontransformer-circuits.pub
• Six (and a half) intuitions for KL divergence — LessWronglesswrong.com
• Thinking about High-Quality Human Data | Lil'Loglilianweng.github.io
• Classificationfairmlbook.org
• When do duplicates/frequencies matter in classification? – Kyunghyun Chokyunghyuncho.me
• 0cfc9404f89400c5ed897035e0d3748c-Paper-Conference.pdfproceedings.neurips.cc
• An Introduction to Exemplar Partitioning for Mechanistic Interpretability — LessWronglesswrong.com
• Musings on typicality – Sander Dielemansander.ai
• Sparse Autoencoders Find Highly Interpretable Features in Language Modelsarxiv.org
• A Bitter Lesson for Data Filteringarxiv.org
• Deep Anomaly Detection with Outlier Exposurearxiv.org
• A Course in Machine Learningciml.info