# L2P: Learning to Place for Estimating Heavy-Tailed Distributed Outcomes

## TL;DR

L2P is a novel method that learns to predict heavy-tailed outcomes by leveraging pairwise comparisons, improving accuracy for rare but significant instances and providing interpretable results.

## Contribution

The paper introduces L2P, a new approach that learns pairwise preferences to better predict heavy-tailed distributions and offers interpretability in predictions.

## Key findings

- L2P outperforms existing methods in accuracy for heavy-tailed data.
- L2P effectively reproduces heavy-tailed outcome distributions.
- L2P provides interpretable placement-based predictions.

## Abstract

Many real-world prediction tasks have outcome variables that have characteristic heavy-tail distributions. Examples include copies of books sold, auction prices of art pieces, demand for commodities in warehouses, etc. By learning heavy-tailed distributions, "big and rare" instances (e.g., the best-sellers) will have accurate predictions. Most existing approaches are not dedicated to learning heavy-tailed distribution; thus, they heavily under-predict such instances. To tackle this problem, we introduce Learning to Place (L2P), which exploits the pairwise relationships between instances for learning. In its training phase, L2P learns a pairwise preference classifier: is instance A > instance B? In its placing phase, L2P obtains a prediction by placing the new instance among the known instances. Based on its placement, the new instance is then assigned a value for its outcome variable. Experiments on real data show that L2P outperforms competing approaches in terms of accuracy and ability to reproduce heavy-tailed outcome distribution. In addition, L2P provides an interpretable model by placing each predicted instance in relation to its comparable neighbors. Interpretable models are highly desirable when lives and treasure are at stake.