Request-Only Optimization for Recommendation Systems

TL;DR

This paper introduces Request-Only Optimization (ROO), a novel paradigm for training recommendation systems that enhances storage efficiency and model quality by treating user requests as the fundamental unit, enabling better data deduplication and scalable architectures.

Contribution

The paper presents a holistic request-only approach that co-designs data, infrastructure, and models to improve efficiency and effectiveness of large-scale recommendation systems.

Findings

Native feature deduplication reduces data storage needs.

De-duplicated computations enable larger, more complex models.

Improved capture of user interest signals with request-only architectures.

Abstract

Deep Learning Recommendation Models (DLRMs) represent one of the largest machine learning applications on the planet. Industry-scale DLRMs are trained with petabytes of recommendation data to serve billions of users every day. To utilize the rich user signals in the long user history, DLRMs have been scaled up to unprecedented complexity, up to trillions of floating-point operations (TFLOPs) per example. This scale, coupled with the huge amount of training data, necessitates new storage and training algorithms to efficiently improve the quality of these complex recommendation systems. In this paper, we present a Request-Only Optimizations (ROO) training and modeling paradigm. ROO simultaneously improves the storage and training efficiency as well as the model quality of recommendation systems. We holistically approach this challenge through co-designing data (i.e., request-only data), infrastructure (i.e., request-only based data processing pipeline), and model architecture (i.e., request-only neural architectures). Our ROO training and modeling paradigm treats a user request as a unit of the training data. Compared with the established practice of treating a user impression as a unit, our new design achieves native feature deduplication in data logging, consequently saving data storage. Second, by de-duplicating computations and communications across multiple impressions in a request, this new paradigm enables highly scaled-up neural network architectures to better capture user interest signals, such as Generative Recommenders (GRs) and other request-only friendly architectures.