TY - GEN
T1 - AutoGen
T2 - 16th ACM International Conference on Web Search and Data Mining (WSDM 2023)
AU - Zhu, Chenxu
AU - Chen, Bo
AU - Guo, Huifeng
AU - Xu, Hang
AU - Li, Xiangyang
AU - Zhao, Xiangyu
AU - Zhang, Weinan
AU - Yu, Yong
AU - Tang, Ruiming
N1 - Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).
PY - 2023
Y1 - 2023
N2 - Considering the balance between revenue and resource consumption for industrial recommender systems, intelligent recommendation computing has been emerging recently. Existing solutions deploy the same recommendation model to serve users indiscriminately, which is sub-optimal for total revenue maximization. We propose a multi-model service solution by deploying different-complexity models to serve different-valued users. An automated dynamic model generation framework AutoGen is elaborated to efficiently derive multiple parameter-sharing models with diverse complexities and adequate predictive capabilities. A mixed search space is designed and an importance-aware progressive training scheme is proposed to prevent interference between different architectures, which avoids the model retraining and improves the search efficiency, thereby efficiently deriving multiple models. Extensive experiments are conducted on two public datasets to demonstrate the effectiveness and efficiency of AutoGen. The code of AutoGen is publicly available.© 2023 Association for Computing Machinery.
AB - Considering the balance between revenue and resource consumption for industrial recommender systems, intelligent recommendation computing has been emerging recently. Existing solutions deploy the same recommendation model to serve users indiscriminately, which is sub-optimal for total revenue maximization. We propose a multi-model service solution by deploying different-complexity models to serve different-valued users. An automated dynamic model generation framework AutoGen is elaborated to efficiently derive multiple parameter-sharing models with diverse complexities and adequate predictive capabilities. A mixed search space is designed and an importance-aware progressive training scheme is proposed to prevent interference between different architectures, which avoids the model retraining and improves the search efficiency, thereby efficiently deriving multiple models. Extensive experiments are conducted on two public datasets to demonstrate the effectiveness and efficiency of AutoGen. The code of AutoGen is publicly available.© 2023 Association for Computing Machinery.
KW - automated machine learning
KW - intelligent computation
KW - neural architecture search
KW - recommender system
UR - http://www.scopus.com/inward/record.url?scp=85149657843&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85149657843&origin=recordpage
U2 - 10.1145/3539597.3570456
DO - 10.1145/3539597.3570456
M3 - RGC 32 - Refereed conference paper (with host publication)
SN - 9781450394079
T3 - WSDM - Proceedings of the ACM International Conference on Web Search and Data Mining
SP - 598
EP - 606
BT - WSDM ’23
PB - Association for Computing Machinery
Y2 - 27 February 2023 through 3 March 2023
ER -