Modeling Orders of User Behaviors via Differentiable Sorting: A Multi-task Framework to Predicting User Post-click Conversion

TL;DR

This paper introduces a multi-task framework that uses differentiable sorting to model the order of user behaviors, improving post-click conversion prediction accuracy by leveraging label relations.

Contribution

It proposes a novel end-to-end multi-task model that incorporates label order information through differentiable sorting, addressing listwise learning in user behavior prediction.

Findings

Outperforms baseline models on public datasets.

Effectively models label relations via differentiable sorting.

Demonstrates industrial applicability with superior results.

Abstract

User post-click conversion prediction is of high interest to researchers and developers. Recent studies employ multi-task learning to tackle the selection bias and data sparsity problem, two severe challenges in post-click behavior prediction, by incorporating click data. However, prior works mainly focused on pointwise learning and the orders of labels (i.e., click and post-click) are not well explored, which naturally poses a listwise learning problem. Inspired by recent advances on differentiable sorting, in this paper, we propose a novel multi-task framework that leverages orders of user behaviors to predict user post-click conversion in an end-to-end approach. Specifically, we define an aggregation operator to combine predicted outputs of different tasks to a unified score, then we use the computed scores to model the label relations via differentiable sorting. Extensive experiments on public and industrial datasets show the superiority of our proposed model against competitive baselines.