Deep Position-wise Interaction Network for CTR Prediction

TL;DR

This paper introduces a Deep Position-wise Interaction Network (DPIN) that models complex interactions between position, user, context, and item to improve CTR prediction accuracy and address position bias, achieving better online performance.

Contribution

The paper proposes a novel DPIN model that captures deep non-linear interactions and ensures consistency between offline training and online inference for CTR prediction.

Findings

Empirically outperforms baseline models on real-world data.

Achieves statistically significant improvements in online A/B testing.

Effectively models position, user, context, and item interactions.

Abstract

Click-through rate (CTR) prediction plays an important role in online advertising and recommender systems. In practice, the training of CTR models depends on click data which is intrinsically biased towards higher positions since higher position has higher CTR by nature. Existing methods such as actual position training with fixed position inference and inverse propensity weighted training with no position inference alleviate the bias problem to some extend. However, the different treatment of position information between training and inference will inevitably lead to inconsistency and sub-optimal online performance. Meanwhile, the basic assumption of these methods, i.e., the click probability is the product of examination probability and relevance probability, is oversimplified and insufficient to model the rich interaction between position and other information. In this paper, we propose a Deep Position-wise Interaction Network (DPIN) to efficiently combine all candidate items and positions for estimating CTR at each position, achieving consistency between offline and online as well as modeling the deep non-linear interaction among position, user, context and item under the limit of serving performance. Following our new treatment to the position bias in CTR prediction, we propose a new evaluation metrics named PAUC (position-wise AUC) that is suitable for measuring the ranking quality at a given position. Through extensive experiments on a real world dataset, we show empirically that our method is both effective and efficient in solving position bias problem. We have also deployed our method in production and observed statistically significant improvement over a highly optimized baseline in a rigorous A/B test.