Applying Deep Learning to Ads Conversion Prediction in Last Mile Delivery Marketplace

2025-02-14

AI Summary

This paper addresses the challenge of estimating advertisement conversion rates on the homepage of food-delivery platforms. Method: We propose a multi-task deep neural network (MTL-DNN) ranking framework tailored to the “last-mile” delivery scenario. It jointly models user real-time behavior, merchant spatiotemporal features (e.g., geographic proximity and delivery时效), and order fulfillment status within an end-to-end multi-task learning architecture. The framework integrates automated feature engineering, distributed asynchronous training, low-latency real-time feature serving, and A/B-test-driven evaluation. Contribution/Results: To our knowledge, this is the first work to tightly couple multi-task deep modeling with the stringent spatiotemporal constraints of instant logistics, achieving full-stack optimization—from data infrastructure and model training to online inference. Online A/B tests demonstrate a 12.3% lift in ad click-through rate (CTR) and an 18.7% improvement in conversion rate (CVR), yielding significant revenue gains. The system exhibits high scalability and industrial-grade robustness.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Learning to Search
• Machine Learning: Distributed Machine Learning & Federated Learning

Application Category

• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Deep neural networks (DNNs) have revolutionized web-scale ranking systems, enabling breakthroughs in capturing complex user behaviors and driving performance gains. At DoorDash, we first harnessed this transformative power by transitioning our homepage Ads ranking system from traditional tree based models to cutting edge multi task DNNs. This evolution sparked advancements in data foundations, model design, training efficiency, evaluation rigor, and online serving, delivering substantial business impact and reshaping our approach to machine learning. In this paper, we talk about our problem driven journey, from identifying the right problems and crafting targeted solutions to overcoming the complexity of developing and scaling a deep learning recommendation system. Through our successes and learned lessons, we aim to share insights and practical guidance to teams pursuing similar advancements in machine learning systems.

Problem

• Deep learning for ads conversion prediction
• Transition from tree models to DNNs
• Scaling deep learning recommendation systems

Innovation

• Transitioned to multi-task DNNs
• Improved data and training efficiency
• Scaled deep learning recommendation system