# Learning Early Exit Strategies for Additive Ranking Ensembles

arXiv: https://arxiv.org/abs/2105.02568 · HTML: https://mlsanity.com/paper/2105.02568/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 May 2021.

## Built on by (most cited first, 1 of 1)

- [Certified Error Control of Candidate Set Pruning for Two-Stage Relevance Ranking](https://mlsanity.com/paper/2205.09638/) (2022)

## Abstract (arXiv)

Modern search engine ranking pipelines are commonly based on large machine-learned ensembles of regression trees. We propose LEAR, a novel - learned - technique aimed to reduce the average number of trees traversed by documents to accumulate the scores, thus reducing the overall query response time. LEAR exploits a classifier that predicts whether a document can early exit the ensemble because it is unlikely to be ranked among the final top-k results. The early exit decision occurs at a sentinel point, i.e., after having evaluated a limited number of trees, and the partial scores are exploited to filter out non-promising documents. We evaluate LEAR by deploying it in a production-like setting, adopting a state-of-the-art algorithm for ensembles traversal. We provide a comprehensive experimental evaluation on two public datasets. The experiments show that LEAR has a significant impact on the efficiency of the query processing without hindering its ranking quality. In detail, on a first dataset, LEAR is able to achieve a speedup of 3x without any loss in NDCG1@0, while on a second dataset the speedup is larger than 5x with a negligible NDCG@10 loss (< 0.05%).