Interaction-Data-guided Conditional Instrumental Variables for Debiasing Recommender Systems
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912550249037824 |
|---|---|
| author | Huang, Zhirong Cheng, Debo Li, Jiuyong Liu, Lin Lu, Guangquan Zhang, Shichao |
| author_facet | Huang, Zhirong Cheng, Debo Li, Jiuyong Liu, Lin Lu, Guangquan Zhang, Shichao |
| contents | It is often challenging to identify a valid instrumental variable (IV), although the IV methods have been regarded as effective tools of addressing the confounding bias introduced by latent variables. To deal with this issue, an Interaction-Data-guided Conditional IV (IDCIV) debiasing method is proposed for Recommender Systems, called IDCIV-RS. The IDCIV-RS automatically generates the representations of valid CIVs and their corresponding conditioning sets directly from interaction data, significantly reducing the complexity of IV selection while effectively mitigating the confounding bias caused by latent variables in recommender systems. Specifically, the IDCIV-RS leverages a variational autoencoder (VAE) to learn both the CIV representations and their conditioning sets from interaction data, followed by the application of least squares to derive causal representations for click prediction. Extensive experiments on two real-world datasets, Movielens-10M and Douban-Movie, demonstrate that IDCIV-RS successfully learns the representations of valid CIVs, effectively reduces bias, and consequently improves recommendation accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_09651 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Interaction-Data-guided Conditional Instrumental Variables for Debiasing Recommender Systems Huang, Zhirong Cheng, Debo Li, Jiuyong Liu, Lin Lu, Guangquan Zhang, Shichao Information Retrieval Artificial Intelligence It is often challenging to identify a valid instrumental variable (IV), although the IV methods have been regarded as effective tools of addressing the confounding bias introduced by latent variables. To deal with this issue, an Interaction-Data-guided Conditional IV (IDCIV) debiasing method is proposed for Recommender Systems, called IDCIV-RS. The IDCIV-RS automatically generates the representations of valid CIVs and their corresponding conditioning sets directly from interaction data, significantly reducing the complexity of IV selection while effectively mitigating the confounding bias caused by latent variables in recommender systems. Specifically, the IDCIV-RS leverages a variational autoencoder (VAE) to learn both the CIV representations and their conditioning sets from interaction data, followed by the application of least squares to derive causal representations for click prediction. Extensive experiments on two real-world datasets, Movielens-10M and Douban-Movie, demonstrate that IDCIV-RS successfully learns the representations of valid CIVs, effectively reduces bias, and consequently improves recommendation accuracy. |
| title | Interaction-Data-guided Conditional Instrumental Variables for Debiasing Recommender Systems |
| topic | Information Retrieval Artificial Intelligence |
| url | https://arxiv.org/abs/2408.09651 |