iHAS: Instance-wise Hierarchical Architecture Search for Deep Learning Recommendation Models

被引:0
|
作者
Yu, Yakun [1 ]
Qi, Shi-Ang [1 ]
Yang, Jiuding [1 ]
Jiang, Liyao [1 ]
Niu, Di [1 ]
机构
[1] Univ Alberta, Edmonton, AB, Canada
关键词
recommender system; instance-wise; embedding dimension search;
D O I
10.1145/3583780.3614925
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Current recommender systems employ large-sized embedding tables with uniform dimensions for all features, leading to overfitting, high computational cost, and suboptimal generalizing performance. Many techniques aim to solve this issue by feature selection or embedding dimension search. However, these techniques typically select a fixed subset of features or embedding dimensions for all instances and feed all instances into one recommender model without considering heterogeneity between items or users. This paper proposes a novel instance-wise Hierarchical Architecture Search framework, iHAS, which automates neural architecture search at the instance level. Specifically, iHAS incorporates three stages: searching, clustering, and retraining. The searching stage identifies optimal instance-wise embedding dimensions across different field features via carefully designed Bernoulli gates with stochastic selection and regularizers. After obtaining these dimensions, the clustering stage divides samples into distinct groups via a deterministic selection approach of Bernoulli gates. The retraining stage then constructs different recommender models, each one designed with optimal dimensions for the corresponding group. We conduct extensive experiments to evaluate the proposed iHAS on two public benchmark datasets from a real-world recommender system. The experimental results demonstrate the effectiveness of iHAS and its outstanding transferability to widely-used deep recommendation models.
引用
收藏
页码:3030 / 3039
页数:10
相关论文
共 50 条
  • [41] Understanding Training Efficiency of Deep Learning Recommendation Models at Scale
    Acun, Bilge
    Murphy, Matthew
    Wang, Xiaodong
    Nie, Jade
    Wu, Carole-Jean
    Hazelwood, Kim
    2021 27TH IEEE INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE COMPUTER ARCHITECTURE (HPCA 2021), 2021, : 802 - 814
  • [42] Learning Hierarchical Features from Deep Generative Models
    Zhao, Shengjia
    Song, Jiaming
    Ermon, Stefano
    INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 70, 2017, 70
  • [43] A Deep Learning Method for Comparing Bayesian Hierarchical Models
    Elsemueller, Lasse
    Schnuerch, Martin
    Buerkner, Paul-Christian
    Radev, Stefan T.
    PSYCHOLOGICAL METHODS, 2024,
  • [44] Step-Wise Deep Learning Models for Solving Routing Problems
    Xin, Liang
    Song, Wen
    Cao, Zhiguang
    Zhang, Jie
    IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2021, 17 (07) : 4861 - 4871
  • [45] SEAnet: A Deep Learning Architecture for Data Series Similarity Search
    Wang, Qitong
    Palpanas, Themis
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2023, 35 (12) : 12972 - 12986
  • [46] Attention-based Deep Reinforcement Learning Model for Pair-Wise Interaction Recommendation
    Zhang, Chenyan
    Guan, Huanmei
    Li, Ni
    2019 6TH INTERNATIONAL CONFERENCE ON INFORMATION SCIENCE AND CONTROL ENGINEERING (ICISCE 2019), 2019, : 169 - 174
  • [47] DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy
    Cheng, Anda
    Wang, Jiaxing
    Zhang, Xi Sheryl
    Chen, Qiang
    Wang, Peisong
    Cheng, Jian
    THIRTY-SIXTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FOURTH CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE / THE TWELVETH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2022, : 6358 - 6366
  • [48] A Hierarchical Deep Video Understanding Method with Shot-Based Instance Search and Large Language Model
    Li, Ruizhe
    Guo, Jiahao
    Li, Mingxi
    Wu, Zhengqian
    Liang, Chao
    PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2023, 2023, : 9425 - 9429
  • [49] Marginal Deep Architecture: Stacking Feature Learning Modules to Build Deep Learning Models
    Zhong, Guoqiang
    Zhang, Kang
    Wei, Hongxu
    Zheng, Yuchen
    Dong, Junyu
    IEEE ACCESS, 2019, 7 : 30220 - 30233
  • [50] Evaluation of Deep Learning Instance Segmentation models for Pig Precision Livestock Farming
    Witte, Jan-Hendrik
    Gerberding, Johann
    Melching, Christian
    Gomez, Jorge Marx
    24TH INTERNATIONAL CONFERENCE ON BUSINESS INFORMATION SYSTEMS (BIS): ENTERPRISE KNOWLEDGE AND DATA SPACES, 2021, : 209 - 220