TwERC: High Performance Ensembled Candidate Generation for Ads Recommendation at Twitter

TL;DR

This paper introduces TwERC, a novel candidate generation system for Twitter ads that combines real-time ranking with sourcing strategies, leading to significant revenue improvements and better understanding of system trade-offs.

Contribution

TwERC presents a new heterogeneous re-architecture for candidate generation that integrates similarity graph and score caching strategies, enhancing ad revenue.

Findings

Graph-based strategy yields 4.08% revenue increase.

Score caching strategy achieves 1.38% revenue gain.

Complementary biases improve overall candidate quality.

Abstract

Recommendation systems are a core feature of social media companies with their uses including recommending organic and promoted contents. Many modern recommendation systems are split into multiple stages - candidate generation and heavy ranking - to balance computational cost against recommendation quality. We focus on the candidate generation phase of a large-scale ads recommendation problem in this paper, and present a machine learning first heterogeneous re-architecture of this stage which we term TwERC. We show that a system that combines a real-time light ranker with sourcing strategies capable of capturing additional information provides validated gains. We present two strategies. The first strategy uses a notion of similarity in the interaction graph, while the second strategy caches previous scores from the ranking stage. The graph based strategy achieves a 4.08% revenue gain and the rankscore based strategy achieves a 1.38% gain. These two strategies have biases that complement both the light ranker and one another. Finally, we describe a set of metrics that we believe are valuable as a means of understanding the complex product trade offs inherent in industrial candidate generation systems.