adSformers: Personalization from Short-Term Sequences and Diversity of Representations in Etsy Ads
Alaa Awad, Denisa Roberts, Eden Dolev, Andrea Heyman, Zahra, Ebrahimzadeh, Zoe Weil, Marcin Mejran, Vaibhav Malpani, Mahir Yavuz

TL;DR
This paper introduces adSformers, a novel personalization module that encodes variable-length user action sequences and diverse representations to improve ad click and conversion predictions, leading to significant offline and online performance gains.
Contribution
The paper presents the adSformer diversifiable personalization module (ADPM), a flexible three-component system that enhances user representations for ad CTR and PCCVR models, outperforming existing baselines.
Findings
ADPM improves CTR and PCCVR models by over 2.4% in offline ROC-AUC.
Deployment of adSformer-based models increased online ad performance.
The approach is effective in personalizing ads using short-term user sequences and diverse data representations.
Abstract
In this article, we present a general approach to personalizing ads through encoding and learning from variable-length sequences of recent user actions and diverse representations. To this end we introduce a three-component module called the adSformer diversifiable personalization module (ADPM) that learns a dynamic user representation. We illustrate the module's effectiveness and flexibility by personalizing the Click-Through Rate (CTR) and Post-Click Conversion Rate (PCCVR) models used in sponsored search. The first component of the ADPM, the adSformer encoder, includes a novel adSformer block which learns the most salient sequence signals. ADPM's second component enriches the learned signal through visual, multimodal, and other pretrained representations. Lastly, the third ADPM "learned on the fly" component further diversifies the signal encoded in the dynamic user representation.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsVideo Analysis and Summarization · Digital Games and Media · Advanced Image and Video Retrieval Techniques
