SEQ+MD: Learning Multi-Task as a SEQuence with Multi-Distribution Data

TL;DR

The paper introduces SEQ+MD, a novel multi-task learning framework that effectively handles multi-distribution data in e-commerce, improving high-value engagement metrics while maintaining click performance.

Contribution

It presents a new sequential multi-task learning approach with feature-generated region-masks for multi-distribution data, addressing regional heterogeneity in e-commerce search ranking.

Findings

Significant increase in add-to-cart and purchase rates.

Maintains neutral click performance compared to baselines.

Plug-and-play multi-regional learning module.

Abstract

In e-commerce, the order in which search results are displayed when a customer tries to find relevant listings can significantly impact their shopping experience and search efficiency. Tailored re-ranking system based on relevance and engagement signals in E-commerce has often shown improvement on sales and gross merchandise value (GMV). Designing algorithms for this purpose is even more challenging when the shops are not restricted to domestic buyers, but can sale globally to international buyers. Our solution needs to incorporate shopping preference and cultural traditions in different buyer markets. We propose the SEQ+MD framework, which integrates sequential learning for multi-task learning (MTL) and feature-generated region-mask for multi-distribution input. This approach leverages the sequential order within tasks and accounts for regional heterogeneity, enhancing performance on multi-source data. Evaluations on in-house data showed a strong increase on the high-value engagement including add-to-cart and purchase while keeping click performance neutral compared to state-of-the-art baseline models. Additionally, our multi-regional learning module is "plug-and-play" and can be easily adapted to enhance other MTL applications.