13.5. 小结

推荐系统作为深度学习在工业界最成功的落地成果之一，极大地提升了用户的在线服务体验，并且为各大公司创造了可观的利润，然而也带来了许多系统层面的挑战亟待解决。本节简单介绍了典型的工业界推荐系统架构及其面临的挑战，并给出了潜在的解决方案的方向。在实际生产环境中，具体的系统设计方案需要根据不同推荐场景的需求而变化，不存在一种万能的解决方案。

13.6. 扩展阅读

13.7. 参考文献

Cheng et al., 2016

Cheng, H.-T., Koc, L., Harmsen, J., Shaked, T., Chandra, T., Aradhye, H., … Shah, H. (2016). Wide & deep learning for recommender systems. Proceedings of the 1st Workshop on Deep Learning for Recommender Systems (pp. 7–10). New York, NY, USA: Association for Computing Machinery. URL: https://doi.org/10.1145/2988450.2988454, doi:10.1145/2988450.2988454

Ginart et al., 2021

Ginart, A., Naumov, M., Mudigere, D., Yang, J., & Zou, J. (2021). Mixed Dimension Embeddings with Application to Memory-Efficient Recommendation Systems.

Gong et al., 2020

Gong, Y., Jiang, Z., Feng, Y., Hu, B., Zhao, K., Liu, Q., & Ou, W. (2020). Edgerec: recommender system on edge in mobile taobao. Proceedings of the 29th ACM International Conference on Information & Knowledge Management (pp. 2477–2484).

Guo et al., 2017

Guo, H., TANG, R., Ye, Y., Li, Z., & He, X. (2017). Deepfm: a factorization-machine based neural network for ctr prediction. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, IJCAI-17 (pp. 1725–1731). URL: https://doi.org/10.24963/ijcai.2017/239, doi:10.24963/ijcai.2017/239

He et al., 2020

He, C., Annavaram, M., & Avestimehr, S. (2020). Larochelle, H., Ranzato, M., Hadsell, R., Balcan, M. F., & Lin, H. (Eds.). Group knowledge transfer: federated learning of large cnns at the edge. Advances in Neural Information Processing Systems (pp. 14068–14080). Curran Associates, Inc. URL: https://proceedings.neurips.cc/paper/2020/file/a1d4c20b182ad7137ab3606f0e3fc8a4-Paper.pdf

Jiang et al., 2021

Jiang, W., He, Z., Zhang, S., Preuß er, T. B., Zeng, K., Feng, L., … Alonso, G. (2021). Smola, A., Dimakis, A., & Stoica, I. (Eds.). Microrec: efficient recommendation inference by hardware and data structure solutions. Proceedings of Machine Learning and Systems (pp. 845–859). URL: https://proceedings.mlsys.org/paper/2021/file/ec8956637a99787bd197eacd77acce5e-Paper.pdf

Naumov et al., 2019

Naumov, M., Mudigere, D., Shi, H.-J. M., Huang, J., Sundaraman, N., Park, J., … others. (2019). Deep learning recommendation model for personalization and recommendation systems. arXiv preprint arXiv:1906.00091.

NVIDIA, 2022

NVIDIA (2022). NVIDIA Merlin. Accessed on 2022-03-24.

Shi et al., 2020

Shi, H.-J. M., Mudigere, D., Naumov, M., & Yang, J. (2020). Compositional embeddings using complementary partitions for memory-efficient recommendation systems. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (pp. 165–175). New York, NY, USA: Association for Computing Machinery. URL: https://doi.org/10.1145/3394486.3403059, doi:10.1145/3394486.3403059

Wang et al., 2017

Wang, R., Fu, B., Fu, G., & Wang, M. (2017). Deep & cross network for ad click predictions. Proceedings of the ADKDD’17. New York, NY, USA: Association for Computing Machinery. URL: https://doi.org/10.1145/3124749.3124754, doi:10.1145/3124749.3124754

Xie et al., 2020

Xie, M., Ren, K., Lu, Y., Yang, G., Xu, Q., Wu, B., … Shu, J. (2020). Kraken: memory-efficient continual learning for large-scale real-time recommendations. SC20: International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 1–17). doi:10.1109/SC41405.2020.00025

Yin et al., 2021

Yin, C., Acun, B., Wu, C.-J., & Liu, X. (2021). Smola, A., Dimakis, A., & Stoica, I. (Eds.). Tt-rec: tensor train compression for deep learning recommendation models. Proceedings of Machine Learning and Systems (pp. 448–462). URL: https://proceedings.mlsys.org/paper/2021/file/979d472a84804b9f647bc185a877a8b5-Paper.pdf

Zhao et al., 2020

Zhao, W., Xie, D., Jia, R., Qian, Y., Ding, R., Sun, M., & Li, P. (2020). Dhillon, I., Papailiopoulos, D., & Sze, V. (Eds.). Distributed hierarchical gpu parameter server for massive scale deep learning ads systems. Proceedings of Machine Learning and Systems (pp. 412–428). URL: https://proceedings.mlsys.org/paper/2020/file/f7e6c85504ce6e82442c770f7c8606f0-Paper.pdf