Task Similarity Aware Meta Learning for Cold-Start Recommendation

Yang, Junyi; Huan, Zhaoxin; He, Yong; Ding, Ke; Zhang, Liang; Zhang, Xiaolu; Zhou, Jun; Mo, Linjian

doi:10.1145/3511808.3557709

Cited by 5 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Three datasets and four models are used for evaluation from two aspects. For statistical correctness, three popular meta DLRM (MAML [35], MeLU [19], and CBML [31]) using Movielens dataset [14] are leveraged for verifying the correctness of the implementation, following the model settings in TSAML [37]. For efficiency, we evaluate G-Meta using an in-house Meta DLRM model on both Ali-CCP dataset [22] and the in-house dataset with 1.6 billion records, since the MovieLens dataset lacks enough samples for large-scale training.…”

Section: Methodsmentioning

confidence: 99%

G-Meta: Distributed Meta Learning in GPU Clusters for Large-Scale Recommender Systems

Xiao,

Zhao,

Zhou

et al. 2023

Proceedings of the 32nd ACM International Conference on Information and Knowledge Management

Self Cite

View full text Add to dashboard Cite

Recently, a new paradigm, meta learning, has been widely applied to Deep Learning Recommendation Models (DLRM) and significantly improves statistical performance, especially in cold-start scenarios. However, the existing systems are not tailored for meta learning based DLRM models and have critical problems regarding efficiency in distributed training in the GPU cluster. It is because the conventional deep learning pipeline is not optimized for two taskspecific datasets and two update loops in meta learning. This paper provides a high-performance framework for large-scale training for Optimization-based Meta DLRM models over the GPU cluster, namely G-Meta. Firstly, G-Meta utilizes both data parallelism and model parallelism with careful orchestration regarding computation and communication efficiency, to enable high-speed distributed training. Secondly, it proposes a Meta-IO pipeline for efficient data ingestion to alleviate the I/O bottleneck. Various experimental results show that G-Meta achieves notable training speed without loss of statistical performance. Since early 2022, G-Meta has been deployed in Alipay's core advertising and recommender system, shrinking the continuous delivery of models by four times. It also obtains 6.48% improvement in Conversion Rate (CVR) and 1.06% increase in CPM (Cost Per Mille) in Alipay's homepage display advertising, with the benefit of larger training samples and tasks.

show abstract