ESCM$^2$: Entire Space Counterfactual Multi-Task Model for Post-Click Conversion Rate Estimation

TL;DR

This paper introduces ESCM$^2$, a novel model that improves post-click conversion rate estimation by addressing bias and causality issues in existing multi-task models, leading to more accurate recommender system predictions.

Contribution

The paper proposes ESCM$^2$, a new counterfactual multi-task model that mitigates inherent bias and causality oversight in existing methods for conversion rate estimation.

Findings

ESCM$^2$ reduces estimation bias compared to baseline models.

The model effectively captures causal relationships in user actions.

Experimental results show improved offline and online performance.

Abstract

Accurate estimation of post-click conversion rate is critical for building recommender systems, which has long been confronted with sample selection bias and data sparsity issues. Methods in the Entire Space Multi-task Model (ESMM) family leverage the sequential pattern of user actions, i.e. $impression\rightarrow click \rightarrow conversion$ to address data sparsity issue. However, they still fail to ensure the unbiasedness of CVR estimates. In this paper, we theoretically demonstrate that ESMM suffers from the following two problems: (1) Inherent Estimation Bias (IEB), where the estimated CVR of ESMM is inherently higher than the ground truth; (2) Potential Independence Priority (PIP) for CTCVR estimation, where there is a risk that the ESMM overlooks the causality from click to conversion. To this end, we devise a principled approach named Entire Space Counterfactual Multi-task Modelling (ESCM$^2$), which employs a counterfactual risk miminizer as a regularizer in ESMM to address both IEB and PIP issues simultaneously. Extensive experiments on offline datasets and online environments demonstrate that our proposed ESCM$^2$ can largely mitigate the inherent IEB and PIP issues and achieve better performance than baseline models.