The Abacus.AI team has published five papers at NeurIPS 2023, including one awarded with an Oral Presentation. They cover topics such as zero-shot forecasting, bias mitigation in face recognition, neural networks vs. boosted trees on tabular data, and improving fairness and accuracy in tabular data models.
Table of contents
ForecastPFN: Synthetically-Trained Zero-Shot ForecastingRethinking Bias Mitigation: Fairer Architectures Make for Fairer Face RecognitionWhen Do Neural Nets Outperform Boosted Trees on Tabular Data?Fairer and More Accurate Models Through NASA Natural Experiment on LLM Data Contamination in Code GenerationSort: