MP2: a momentum contrast approach for recommendation with pointwise and pairwise learning
Binary pointwise labels (aka implicit feedback) are heavily leveraged by deep learning based recommendation algorithms nowadays. In this paper we discuss the limited expressiveness of these labels may fail to accommodate varying degrees of user preference, and thus lead to conflicts during model tra...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
Association for Computing Machinery
2022
|
| _version_ | 1797107999497519104 |
|---|---|
| author | Wang, M Guo, Y Zhao, Z Hu, G Shen, Y Gong, M Torr, P |
| author_facet | Wang, M Guo, Y Zhao, Z Hu, G Shen, Y Gong, M Torr, P |
| author_sort | Wang, M |
| collection | OXFORD |
| description | Binary pointwise labels (aka implicit feedback) are heavily leveraged by deep learning based recommendation algorithms nowadays. In this paper we discuss the limited expressiveness of these labels may fail to accommodate varying degrees of user preference, and thus lead to conflicts during model training, which we call annotation bias. To solve this issue, we find the soft-labeling property of pairwise labels could be utilized to alleviate the bias of pointwise labels. To this end, we propose a momentum contrast framework (\method ) that combines pointwise and pairwise learning for recommendation. \method has a three-tower network structure: one user network and two item networks. The two item networks are used for computing pointwise and pairwise loss respectively. To alleviate the influence of the annotation bias, we perform a momentum update to ensure a consistent item representation. Extensive experiments on real-world datasets demonstrate the superiority of our method against state-of-the-art recommendation algorithms. |
| first_indexed | 2024-03-07T07:23:22Z |
| format | Conference item |
| id | oxford-uuid:c1da435c-b57f-4614-88bb-53cccfc8ae9c |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:23:22Z |
| publishDate | 2022 |
| publisher | Association for Computing Machinery |
| record_format | dspace |
| spelling | oxford-uuid:c1da435c-b57f-4614-88bb-53cccfc8ae9c2022-11-01T07:34:46ZMP2: a momentum contrast approach for recommendation with pointwise and pairwise learningConference itemhttp://purl.org/coar/resource_type/c_5794uuid:c1da435c-b57f-4614-88bb-53cccfc8ae9cEnglishSymplectic ElementsAssociation for Computing Machinery2022Wang, MGuo, YZhao, ZHu, GShen, YGong, MTorr, PBinary pointwise labels (aka implicit feedback) are heavily leveraged by deep learning based recommendation algorithms nowadays. In this paper we discuss the limited expressiveness of these labels may fail to accommodate varying degrees of user preference, and thus lead to conflicts during model training, which we call annotation bias. To solve this issue, we find the soft-labeling property of pairwise labels could be utilized to alleviate the bias of pointwise labels. To this end, we propose a momentum contrast framework (\method ) that combines pointwise and pairwise learning for recommendation. \method has a three-tower network structure: one user network and two item networks. The two item networks are used for computing pointwise and pairwise loss respectively. To alleviate the influence of the annotation bias, we perform a momentum update to ensure a consistent item representation. Extensive experiments on real-world datasets demonstrate the superiority of our method against state-of-the-art recommendation algorithms. |
| spellingShingle | Wang, M Guo, Y Zhao, Z Hu, G Shen, Y Gong, M Torr, P MP2: a momentum contrast approach for recommendation with pointwise and pairwise learning |
| title | MP2: a momentum contrast approach for recommendation with pointwise and pairwise learning |
| title_full | MP2: a momentum contrast approach for recommendation with pointwise and pairwise learning |
| title_fullStr | MP2: a momentum contrast approach for recommendation with pointwise and pairwise learning |
| title_full_unstemmed | MP2: a momentum contrast approach for recommendation with pointwise and pairwise learning |
| title_short | MP2: a momentum contrast approach for recommendation with pointwise and pairwise learning |
| title_sort | mp2 a momentum contrast approach for recommendation with pointwise and pairwise learning |
| work_keys_str_mv | AT wangm mp2amomentumcontrastapproachforrecommendationwithpointwiseandpairwiselearning AT guoy mp2amomentumcontrastapproachforrecommendationwithpointwiseandpairwiselearning AT zhaoz mp2amomentumcontrastapproachforrecommendationwithpointwiseandpairwiselearning AT hug mp2amomentumcontrastapproachforrecommendationwithpointwiseandpairwiselearning AT sheny mp2amomentumcontrastapproachforrecommendationwithpointwiseandpairwiselearning AT gongm mp2amomentumcontrastapproachforrecommendationwithpointwiseandpairwiselearning AT torrp mp2amomentumcontrastapproachforrecommendationwithpointwiseandpairwiselearning |