Summary of Mixture Of In-context Prompters For Tabular Pfns, by Derek Xu et al.
Mixture of In-Context Prompters for Tabular PFNs
by Derek Xu, Olcay Cirit, Reza Asadi, Yizhou Sun, Wei Wang
First submitted to arxiv on: 25 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary In a recent study, researchers found that In-Context Learning (ICL) outperforms both deep learning and tree-based algorithms on small tabular datasets. However, ICL’s limitations in handling larger datasets are significant, as it suffers from quadratic space and time complexity issues. To address this challenge, the authors propose MIXTUREPFN, an innovative approach that combines nearest-neighbor sampling with bootstrapping to fine-tune the model during inference. MIXTUREPFN emerges as the top-performing algorithm across 36 diverse tabular datasets, outperforming 19 strong baselines from deep learning and tree-based methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study shows how In-Context Learning can be very good at solving problems on small datasets, but it gets stuck when dealing with larger ones. To fix this problem, scientists created a new way to use ICL called MIXTUREPFN. It works by sampling nearby data points and then adjusting the model based on that information. This approach was tested on many different datasets and outperformed other popular methods. |
Keywords
* Artificial intelligence * Bootstrapping * Deep learning * Inference * Nearest neighbor
