Distribution-Aware End-to-End Embedding for Streaming Numerical Features in Click-Through Rate Prediction

TL;DR

This paper introduces DAES, a novel end-to-end framework for embedding streaming numerical features in CTR prediction, effectively capturing distributional information and adapting to non-i.i.d. streaming data.

Contribution

The paper proposes DAES, which integrates distribution estimation and modulation strategies into neural embeddings for streaming numerical features, addressing semantic drift and distributional shifts.

Findings

DAES outperforms existing methods in offline and online tests.

Successfully deployed on a large-scale short-video platform.

Improves embedding quality for streaming numerical features.

Abstract

This paper explores effective numerical feature embedding for Click-Through Rate prediction in streaming environments. Conventional static binning methods rely on offline statistics of numerical distributions; however, this inherently two-stage process often triggers semantic drift during bin boundary updates. While neural embedding methods enable end-to-end learning, they often discard explicit distributional information. Integrating such information end-to-end is challenging because streaming features often violate the i.i.d. assumption, precluding unbiased estimation of the population distribution via the expectation of order statistics. Furthermore, the critical context dependency of numerical distributions is often neglected. To this end, we propose DAES, an end-to-end framework designed to tackle numerical feature embedding in streaming training scenarios by integrating distributional information with an adaptive modulation mechanism. Specifically, we introduce an efficient reservoir-sampling-based distribution estimation method and two field-aware distribution modulation strategies to capture streaming distributions and field-dependent semantics. DAES significantly outperforms existing approaches as demonstrated by extensive offline and online experiments and has been fully deployed on a leading short-video platform with hundreds of millions of daily active users.