Off-Policy Evaluation for Ranking Policies under Deterministic Logging Policies

TL;DR

This paper introduces Click-based Inverse Propensity Score (CIPS), a novel off-policy evaluation method that leverages user click behavior to accurately estimate ranking policy performance even with deterministic logging policies.

Contribution

The paper proposes CIPS, a new estimator that uses click probabilities to enable low-bias off-policy evaluation under deterministic logging policies, overcoming limitations of existing methods.

Findings

CIPS achieves significantly lower bias than baseline estimators.

Theoretical analysis confirms favorable bias and variance properties.

Experimental results demonstrate effectiveness in real-world and synthetic data.

Abstract

Off-Policy Evaluation (OPE) is an important practical problem in algorithmic ranking systems, where the goal is to estimate the expected performance of a new ranking policy using only offline logged data collected under a different, logging policy. Existing estimators, such as the ranking-wise and position-wise inverse propensity score (IPS) estimators, require the data collection policy to be sufficiently stochastic and suffer from severe bias when the logging policy is fully deterministic. In this paper, we propose novel estimators, Click-based Inverse Propensity Score (CIPS), exploiting the intrinsic stochasticity of user click behavior to address this challenge. Unlike existing methods that rely on the stochasticity of the logging policy, our approach uses click probability as a new form of importance weighting, enabling low-bias OPE even under deterministic logging policies where existing methods incur substantial bias. We provide theoretical analyses of the bias and variance properties of the proposed estimators and show, through synthetic and real-world experiments, that our estimators achieve significantly lower bias compared to strong baselines, for a range of experimental settings with completely deterministic logging policies.