Difference between revisions of "Meta"

From RecSysWiki
Jump to navigation Jump to search
 
(4 intermediate revisions by the same user not shown)
Line 3: Line 3:
 
== Blog posts ==
 
== Blog posts ==
  
 +
# [https://engineering.fb.com/2022/10/31/ml-applications/instagram-notification-management-machine-learning/ Improving Instagram notification management with machine learning and causal inference], 2022-10-31
 
# [https://ai.facebook.com/blog/facebook-feed-improvements-ai-show-more-less/ The new AI-powered feature designed to improve Feed for everyone], 2022-10-05
 
# [https://ai.facebook.com/blog/facebook-feed-improvements-ai-show-more-less/ The new AI-powered feature designed to improve Feed for everyone], 2022-10-05
 +
# [https://ai.facebook.com/tools/system-cards/instagram-feed-ranking/# What is the Instagram Feed?], 2022-02-23
 
# [https://research.fb.com/blog/2021/08/when-do-recommender-systems-amplify-user-preferences-a-theoretical-framework-and-mitigation-strategies/ When do recommender systems amplify user preferences? A theoretical framework and mitigation strategies], 2021-08
 
# [https://research.fb.com/blog/2021/08/when-do-recommender-systems-amplify-user-preferences-a-theoretical-framework-and-mitigation-strategies/ When do recommender systems amplify user preferences? A theoretical framework and mitigation strategies], 2021-08
−
# [https://research.fb.com/efficient-tuning-of-online-systems-using-bayesian-optimization/ Efficient tuning of online systems using Bayesian optimization]
+
# [https://about.instagram.com/blog/announcements/shedding-more-light-on-how-instagram-works Shedding More Light on How Instagram Works], 2021-06-08
 
# [https://engineering.fb.com/2020/12/17/ml-applications/diversified-recommendations/ On the value of diversified recommendations], 2020-12-17
 
# [https://engineering.fb.com/2020/12/17/ml-applications/diversified-recommendations/ On the value of diversified recommendations], 2020-12-17
 
# [https://engineering.fb.com/2020/12/10/web/how-instagram-suggests-new-content/ How Instagram suggests new content], 2020-12-10
 
# [https://engineering.fb.com/2020/12/10/web/how-instagram-suggests-new-content/ How Instagram suggests new content], 2020-12-10
 +
# [https://about.instagram.com/blog/engineering/designing-a-constrained-exploration-system Designing a Constrained Exploration System], 2020-12-10
 
# [https://instagram-engineering.com/five-things-i-learned-about-working-on-content-quality-at-instagram-5031b1342bea Five things I learned about working on content quality at Instagram], 2020-01-25
 
# [https://instagram-engineering.com/five-things-i-learned-about-working-on-content-quality-at-instagram-5031b1342bea Five things I learned about working on content quality at Instagram], 2020-01-25
 
# {{production}} [https://instagram-engineering.com/powered-by-ai-instagrams-explore-recommender-system-7ca901d2a882 Instagram’s Explore Recommender System], 2019-11-26 [https://news.ycombinator.com/item?id=21638216 HackerNews discussion]
 
# {{production}} [https://instagram-engineering.com/powered-by-ai-instagrams-explore-recommender-system-7ca901d2a882 Instagram’s Explore Recommender System], 2019-11-26 [https://news.ycombinator.com/item?id=21638216 HackerNews discussion]
Line 17: Line 20:
 
# [https://edoconti.medium.com/offline-policy-evaluation-run-fewer-better-a-b-tests-60ce8f93fa15 Offline Policy Evaluation: Run fewer, better A/B tests]
 
# [https://edoconti.medium.com/offline-policy-evaluation-run-fewer-better-a-b-tests-60ce8f93fa15 Offline Policy Evaluation: Run fewer, better A/B tests]
 
# [https://ai.facebook.com/blog/dlrm-an-advanced-open-source-deep-learning-recommendation-model/ DLRM: An advanced, open source deep learning recommendation model], 2019-07-02
 
# [https://ai.facebook.com/blog/dlrm-an-advanced-open-source-deep-learning-recommendation-model/ DLRM: An advanced, open source deep learning recommendation model], 2019-07-02
 +
# [https://instagram-engineering.com/lessons-learned-at-instagram-stories-and-feed-machine-learning-54f3aaa09e56 Lessons Learned at Instagram Stories and Feed Machine Learning], 2018-12-18
 +
# [https://research.facebook.com/blog/2018/9/efficient-tuning-of-online-systems-using-bayesian-optimization/ Efficient tuning of online systems using Bayesian optimization], 2018-09-17
  
 
== Papers ==
 
== Papers ==
Line 22: Line 27:
 
# {{ops}}{{performance}} [https://arxiv.org/pdf/2104.05158.pdf Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models], ISCA 2022 industry track
 
# {{ops}}{{performance}} [https://arxiv.org/pdf/2104.05158.pdf Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models], ISCA 2022 industry track
 
# {{ops}}{{performance}} [https://dl.acm.org/doi/abs/10.1145/3534678.3539034 AutoShard: Automated Embedding Table Sharding for Recommender Systems], [[KDD 2022]]
 
# {{ops}}{{performance}} [https://dl.acm.org/doi/abs/10.1145/3534678.3539034 AutoShard: Automated Embedding Table Sharding for Recommender Systems], [[KDD 2022]]
−
# [https://arxiv.org/pdf/2203.11014.pdf DHEN: A Deep and Hierarchical Ensemble Network for Large-Scale Click-Through Rate Prediction], KDD 2022
+
# {{ads}}{{neural}} [https://arxiv.org/pdf/2203.11014.pdf DHEN: A Deep and Hierarchical Ensemble Network for Large-Scale Click-Through Rate Prediction], KDD 2022
 
# https://research.fb.com/publications/the-decoupled-extended-kalman-filter-for-dynamic-exponential-family-factorization-models/, [[JMLR]], 2021
 
# https://research.fb.com/publications/the-decoupled-extended-kalman-filter-for-dynamic-exponential-family-factorization-models/, [[JMLR]], 2021
 
# [https://research.fb.com/publications/preference-amplification-in-recommender-systems/ Preference Amplification in Recommender Systems], [[KDD 2021]]
 
# [https://research.fb.com/publications/preference-amplification-in-recommender-systems/ Preference Amplification in Recommender Systems], [[KDD 2021]]
Line 28: Line 33:
 
# [https://www.kdd.org/kdd2020/accepted-papers/view/compositional-embeddings-using-complementary-partitions-for-memory-efficien Compositional Embeddings Using Complementary Partitions for Memory-Efficient Recommendation Systems], [[KDD 2020]]
 
# [https://www.kdd.org/kdd2020/accepted-papers/view/compositional-embeddings-using-complementary-partitions-for-memory-efficien Compositional Embeddings Using Complementary Partitions for Memory-Efficient Recommendation Systems], [[KDD 2020]]
 
# {{fashion}} [https://arxiv.org/abs/2011.09663 Modeling Fashion Influence from Photos], IEEE Transactions on Multimedia, 2020
 
# {{fashion}} [https://arxiv.org/abs/2011.09663 Modeling Fashion Influence from Photos], IEEE Transactions on Multimedia, 2020
 +
# {{experimentation}} [https://dominiccoey.github.io/assets/papers/experiment_splitting.pdf Improving Treatment Effect Estimators Through Experiment Splitting], WWW 2019
 
# {{neural}} [https://arxiv.org/pdf/1906.00091.pdf Deep Learning Recommendation Model for Personalization and Recommendation Systems], 2019 (DLRM)
 
# {{neural}} [https://arxiv.org/pdf/1906.00091.pdf Deep Learning Recommendation Model for Personalization and Recommendation Systems], 2019 (DLRM)
 
# {{ads}} [https://research.fb.com/publications/counterfactual-reasoning-and-learning-systems-the-example-of-computational-advertising/ Counterfactual Reasoning and Learning Systems: The Example of Computational Advertising]
 
# {{ads}} [https://research.fb.com/publications/counterfactual-reasoning-and-learning-systems-the-example-of-computational-advertising/ Counterfactual Reasoning and Learning Systems: The Example of Computational Advertising]
Line 36: Line 42:
 
== Software ==
 
== Software ==
  
 +
# [https://github.com/pytorch/torchrec TorchRec], [https://dl.acm.org/doi/10.1145/3523227.3547387 RecSys 2022 talk]
 
# [https://github.com/facebookresearch/dlrm DLRM recommender]: click probability model
 
# [https://github.com/facebookresearch/dlrm DLRM recommender]: click probability model
 
# [https://github.com/facebook/prophet Prophet]: forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects.
 
# [https://github.com/facebook/prophet Prophet]: forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects.

Latest revision as of 03:49, 9 April 2024

Meta is the company behind Facebook, Instagram, and WhatsApp.

Blog posts

  1. Improving Instagram notification management with machine learning and causal inference, 2022-10-31
  2. The new AI-powered feature designed to improve Feed for everyone, 2022-10-05
  3. What is the Instagram Feed?, 2022-02-23
  4. When do recommender systems amplify user preferences? A theoretical framework and mitigation strategies, 2021-08
  5. Shedding More Light on How Instagram Works, 2021-06-08
  6. On the value of diversified recommendations, 2020-12-17
  7. How Instagram suggests new content, 2020-12-10
  8. Designing a Constrained Exploration System, 2020-12-10
  9. Five things I learned about working on content quality at Instagram, 2020-01-25
  10. 🕴 Instagram’s Explore Recommender System, 2019-11-26 HackerNews discussion
    • 3-part funnel (2 layers of candidate generation)
    • domain-specific language (We have separation between model and filter config).
    • account embeddings
    • embedding-based
    • “See fewer posts like this” – explicit feedback
  11. Offline Policy Evaluation: Run fewer, better A/B tests
  12. DLRM: An advanced, open source deep learning recommendation model, 2019-07-02
  13. Lessons Learned at Instagram Stories and Feed Machine Learning, 2018-12-18
  14. Efficient tuning of online systems using Bayesian optimization, 2018-09-17

Papers

  1. 🔩🏃 Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Models, ISCA 2022 industry track
  2. 🔩🏃 AutoShard: Automated Embedding Table Sharding for Recommender Systems, KDD 2022
  3. 💵🧠 DHEN: A Deep and Hierarchical Ensemble Network for Large-Scale Click-Through Rate Prediction, KDD 2022
  4. https://research.fb.com/publications/the-decoupled-extended-kalman-filter-for-dynamic-exponential-family-factorization-models/, JMLR, 2021
  5. Preference Amplification in Recommender Systems, KDD 2021
  6. 🔍 Embedding-based Retrieval in Facebook Search, KDD 2020
  7. Compositional Embeddings Using Complementary Partitions for Memory-Efficient Recommendation Systems, KDD 2020
  8. 👗 Modeling Fashion Influence from Photos, IEEE Transactions on Multimedia, 2020
  9. 🔬 Improving Treatment Effect Estimators Through Experiment Splitting, WWW 2019
  10. 🧠 Deep Learning Recommendation Model for Personalization and Recommendation Systems, 2019 (DLRM)
  11. 💵 Counterfactual Reasoning and Learning Systems: The Example of Computational Advertising
  12. Mixed Dimension Embeddings with Application to Memory-Efficient Recommendation Systems
  13. 💵 Practical lessons from predicting clicks on ads at facebook, workshop 2014
  14. Supervised Random Walks: Predicting and Recommending Links in Social Networks, WSDM 2011

Software

  1. TorchRec, RecSys 2022 talk
  2. DLRM recommender: click probability model
  3. Prophet: forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects.
  4. Facebook AI Performance Evaluation Platform: framework and backend agnostic benchmarking platform to compare machine learning inferencing runtime metrics on a set of models and on variety of backends.
  5. 📑 StarSpace: Learning embeddings for classification, retrieval and ranking.
  6. ReAgent/Horizon: end-to-end platform designed to solve industry applied RL problems where datasets are large (millions to billions of observations), the feedback loop is slow (vs. a simulator), and experiments must be done with care because they don’t run in a simulator. Tutorial contains e-commerce/recommendation example. paper
  7. faiss: library for efficient similarity search and clustering of dense vectors.
  8. pysparnn: approximate nearest neighbor search for sparse data in Python.
  9. Ax: adaptive experimentation platform, ax.dev

External links

Retrieved from "https://recsyswiki.com/index.php?title=Meta&oldid=2697"