{ "id": 366704, "title": "The World of RecSys", "url": "/competitions/otto-recommender-system/discussion/366704", "author_name": "Kirderf", "post_date": "2022-11-17T09:29:27.907000", "votes": 10, "comment_count": 0, "views": 0, "content": "

Below are some interesting ideas, SOTA and knowledge in the area of recommender systems.
\nI'll keep it updated with interesting content when more is found along the way.

\n

Personally I will take a deeper look at the latest in AutoRecSys and RL-RecSys, the knowledge one collect is half of the value of the competitions.

\n

Frameworks

\n

TensorFlow Recommenders
\nTensorFlow Recommenders is a library for building recommender system models using TensorFlow.
\nhttps://github.com/tensorflow/recommenders

\n

TorchRec (Beta Release)
\nTorchRec is a PyTorch domain library built to provide common sparsity & parallelism primitives needed for large-scale recommender systems (RecSys). It allows authors to train models with large embedding tables sharded across many GPUs.
\nhttps://github.com/pytorch/torchrec

\n

SlateQ - RL for recommender systems
\nhttps://docs.ray.io/en/master/rllib/rllib-algorithms.html#slateq
\nhttps://github.com/ray-project/ray/tree/master/rllib

\n

AutoRec
\nAutoRec is a Keras-based implementation of automated recommendation algorithms for both rating prediction and Click Through Rate task.
\nhttps://github.com/datamllab/AutoRec

\n

Recmetrics
\nA python library of evalulation metrics and diagnostic tools for recommender systems.
\nhttps://github.com/statisticianinstilettos/recmetrics

\n

Case Recommender
\nCase Recommender is a Python implementation of a number of popular recommendation algorithms for both implicit and explicit feedback.
\nhttps://github.com/caserec/CaseRecommender

\n

Auto-CaseRec
\nAuto-CaseRec: Automatically Selecting and Optimizing Recommendation-Systems Algorithms
\nhttps://www.researchgate.net/publication/345922486_Auto-CaseRec_Automatically_Selecting_and_Optimizing_Recommendation-Systems_Algorithms
\nhttps://github.com/srijang97/Auto-CaseRec

\n

Surprise
\nSurprise is a Python scikit for building and analyzing recommender systems that deal with explicit rating data.
\nhttps://surpriselib.com/

\n

Auto-Surprise
\nAuto-Surprise is built as a wrapper around the Python Surprise recommender-system library. It automates algorithm selection and hyper parameter optimization in a highly parallelized manner.
\nhttps://github.com/ISG-Siegen/Auto-Surprise

\n

Papers and other

\n

Zero-Shot Recommender Systems
\nhttps://arxiv.org/abs/2105.08318

\n

Graph Learning based Recommender Systems
\nhttps://arxiv.org/abs/2105.06339

\n

AutoML for Deep Recommender Systems: A Survey
\nhttps://arxiv.org/abs/2203.13922

\n

An Overview of Recommender Systems and Machine Learning in Feature Modeling and Configuration
\nhttps://arxiv.org/abs/2102.06634

\n

A Survey of Recommender System Techniques and the Ecommerce Domain
\nhttps://arxiv.org/abs/2208.07399

\n

EvalRS: a Rounded Evaluation of Recommender Systems
\nhttps://arxiv.org/abs/2207.05772

\n

Behavior Sequence Transformer for E-commerce Recommendation in Alibaba
\nhttps://arxiv.org/abs/1905.06874
\nhttps://keras.io/examples/structured_data/movielens_recommendations_transformers/

\n

ACM Conference on Recommender Systems Years 2007-2022
\nWith for example below from RECSYS 2022 (SEATTLE) tutorials and ideas for further reading.
\nhttps://recsys.acm.org/recsys22/tutorials/

\n", "messages": [ { "id": 2033473, "postDate": "2022-11-17T09:29:27.907Z", "content": "

Below are some interesting ideas, SOTA and knowledge in the area of recommender systems.
\nI'll keep it updated with interesting content when more is found along the way.

\n

Personally I will take a deeper look at the latest in AutoRecSys and RL-RecSys, the knowledge one collect is half of the value of the competitions.

\n

Frameworks

\n

TensorFlow Recommenders
\nTensorFlow Recommenders is a library for building recommender system models using TensorFlow.
\nhttps://github.com/tensorflow/recommenders

\n

TorchRec (Beta Release)
\nTorchRec is a PyTorch domain library built to provide common sparsity & parallelism primitives needed for large-scale recommender systems (RecSys). It allows authors to train models with large embedding tables sharded across many GPUs.
\nhttps://github.com/pytorch/torchrec

\n

SlateQ - RL for recommender systems
\nhttps://docs.ray.io/en/master/rllib/rllib-algorithms.html#slateq
\nhttps://github.com/ray-project/ray/tree/master/rllib

\n

AutoRec
\nAutoRec is a Keras-based implementation of automated recommendation algorithms for both rating prediction and Click Through Rate task.
\nhttps://github.com/datamllab/AutoRec

\n

Recmetrics
\nA python library of evalulation metrics and diagnostic tools for recommender systems.
\nhttps://github.com/statisticianinstilettos/recmetrics

\n

Case Recommender
\nCase Recommender is a Python implementation of a number of popular recommendation algorithms for both implicit and explicit feedback.
\nhttps://github.com/caserec/CaseRecommender

\n

Auto-CaseRec
\nAuto-CaseRec: Automatically Selecting and Optimizing Recommendation-Systems Algorithms
\nhttps://www.researchgate.net/publication/345922486_Auto-CaseRec_Automatically_Selecting_and_Optimizing_Recommendation-Systems_Algorithms
\nhttps://github.com/srijang97/Auto-CaseRec

\n

Surprise
\nSurprise is a Python scikit for building and analyzing recommender systems that deal with explicit rating data.
\nhttps://surpriselib.com/

\n

Auto-Surprise
\nAuto-Surprise is built as a wrapper around the Python Surprise recommender-system library. It automates algorithm selection and hyper parameter optimization in a highly parallelized manner.
\nhttps://github.com/ISG-Siegen/Auto-Surprise

\n

Papers and other

\n

Zero-Shot Recommender Systems
\nhttps://arxiv.org/abs/2105.08318

\n

Graph Learning based Recommender Systems
\nhttps://arxiv.org/abs/2105.06339

\n

AutoML for Deep Recommender Systems: A Survey
\nhttps://arxiv.org/abs/2203.13922

\n

An Overview of Recommender Systems and Machine Learning in Feature Modeling and Configuration
\nhttps://arxiv.org/abs/2102.06634

\n

A Survey of Recommender System Techniques and the Ecommerce Domain
\nhttps://arxiv.org/abs/2208.07399

\n

EvalRS: a Rounded Evaluation of Recommender Systems
\nhttps://arxiv.org/abs/2207.05772

\n

Behavior Sequence Transformer for E-commerce Recommendation in Alibaba
\nhttps://arxiv.org/abs/1905.06874
\nhttps://keras.io/examples/structured_data/movielens_recommendations_transformers/

\n

ACM Conference on Recommender Systems Years 2007-2022
\nWith for example below from RECSYS 2022 (SEATTLE) tutorials and ideas for further reading.
\nhttps://recsys.acm.org/recsys22/tutorials/

\n", "rawMarkdown": "Below are some interesting ideas, SOTA and knowledge in the area of recommender systems.\nI'll keep it updated with interesting content when more is found along the way.\n\nPersonally I will take a deeper look at the latest in AutoRecSys and RL-RecSys, the knowledge one collect is half of the value of the competitions.\n\n**Frameworks**\n\nTensorFlow Recommenders\nTensorFlow Recommenders is a library for building recommender system models using TensorFlow.\nhttps://github.com/tensorflow/recommenders\n\nTorchRec (Beta Release)\nTorchRec is a PyTorch domain library built to provide common sparsity & parallelism primitives needed for large-scale recommender systems (RecSys). It allows authors to train models with large embedding tables sharded across many GPUs.\nhttps://github.com/pytorch/torchrec\n\nSlateQ - RL for recommender systems\nhttps://docs.ray.io/en/master/rllib/rllib-algorithms.html#slateq\nhttps://github.com/ray-project/ray/tree/master/rllib\n\nAutoRec\nAutoRec is a Keras-based implementation of automated recommendation algorithms for both rating prediction and Click Through Rate task.\nhttps://github.com/datamllab/AutoRec\n\nRecmetrics\nA python library of evalulation metrics and diagnostic tools for recommender systems.\nhttps://github.com/statisticianinstilettos/recmetrics\n\nCase Recommender\nCase Recommender is a Python implementation of a number of popular recommendation algorithms for both implicit and explicit feedback. \nhttps://github.com/caserec/CaseRecommender\n\nAuto-CaseRec\nAuto-CaseRec: Automatically Selecting and Optimizing Recommendation-Systems Algorithms\nhttps://www.researchgate.net/publication/345922486_Auto-CaseRec_Automatically_Selecting_and_Optimizing_Recommendation-Systems_Algorithms\nhttps://github.com/srijang97/Auto-CaseRec\n\nSurprise \nSurprise is a Python scikit for building and analyzing recommender systems that deal with explicit rating data.\nhttps://surpriselib.com/\n\nAuto-Surprise\nAuto-Surprise is built as a wrapper around the Python Surprise recommender-system library. It automates algorithm selection and hyper parameter optimization in a highly parallelized manner.\nhttps://github.com/ISG-Siegen/Auto-Surprise\n\n**Papers and other**\n\nZero-Shot Recommender Systems\nhttps://arxiv.org/abs/2105.08318\n\nGraph Learning based Recommender Systems\nhttps://arxiv.org/abs/2105.06339\n\nAutoML for Deep Recommender Systems: A Survey\nhttps://arxiv.org/abs/2203.13922\n\nAn Overview of Recommender Systems and Machine Learning in Feature Modeling and Configuration\nhttps://arxiv.org/abs/2102.06634\n\nA Survey of Recommender System Techniques and the Ecommerce Domain\nhttps://arxiv.org/abs/2208.07399\n\nEvalRS: a Rounded Evaluation of Recommender Systems\nhttps://arxiv.org/abs/2207.05772\n\nBehavior Sequence Transformer for E-commerce Recommendation in Alibaba\nhttps://arxiv.org/abs/1905.06874\nhttps://keras.io/examples/structured_data/movielens_recommendations_transformers/\n\nACM Conference on Recommender Systems Years 2007-2022\nWith for example below from RECSYS 2022 (SEATTLE) tutorials and ideas for further reading.\nhttps://recsys.acm.org/recsys22/tutorials/\n- Neural Re-ranking for Multi-stage Recommender Systems.\n- Hands-on Reinforcement learning for recommender systems – From Bandits to SlateQ to Offline RL with Ray RLlib.\n- Offline Evaluation for Group Recommender Systems.\n- Training and Deploying Multi-Stage Recommender Systems.\n- Improving Recommender Systems with Human-in-the-Loop.\n- Hands on Explainable Recommender Systems with Knowledge Graphs.\n- Psychology-informed Recommender Systems.\n- Conversational Recommender System Using Deep Reinforcement Learning.\n", "votes": 10 } ], "comments": [], "raw_markdown_by_id": { "2033473": "Below are some interesting ideas, SOTA and knowledge in the area of recommender systems.\nI'll keep it updated with interesting content when more is found along the way.\n\nPersonally I will take a deeper look at the latest in AutoRecSys and RL-RecSys, the knowledge one collect is half of the value of the competitions.\n\n**Frameworks**\n\nTensorFlow Recommenders\nTensorFlow Recommenders is a library for building recommender system models using TensorFlow.\nhttps://github.com/tensorflow/recommenders\n\nTorchRec (Beta Release)\nTorchRec is a PyTorch domain library built to provide common sparsity & parallelism primitives needed for large-scale recommender systems (RecSys). It allows authors to train models with large embedding tables sharded across many GPUs.\nhttps://github.com/pytorch/torchrec\n\nSlateQ - RL for recommender systems\nhttps://docs.ray.io/en/master/rllib/rllib-algorithms.html#slateq\nhttps://github.com/ray-project/ray/tree/master/rllib\n\nAutoRec\nAutoRec is a Keras-based implementation of automated recommendation algorithms for both rating prediction and Click Through Rate task.\nhttps://github.com/datamllab/AutoRec\n\nRecmetrics\nA python library of evalulation metrics and diagnostic tools for recommender systems.\nhttps://github.com/statisticianinstilettos/recmetrics\n\nCase Recommender\nCase Recommender is a Python implementation of a number of popular recommendation algorithms for both implicit and explicit feedback. \nhttps://github.com/caserec/CaseRecommender\n\nAuto-CaseRec\nAuto-CaseRec: Automatically Selecting and Optimizing Recommendation-Systems Algorithms\nhttps://www.researchgate.net/publication/345922486_Auto-CaseRec_Automatically_Selecting_and_Optimizing_Recommendation-Systems_Algorithms\nhttps://github.com/srijang97/Auto-CaseRec\n\nSurprise \nSurprise is a Python scikit for building and analyzing recommender systems that deal with explicit rating data.\nhttps://surpriselib.com/\n\nAuto-Surprise\nAuto-Surprise is built as a wrapper around the Python Surprise recommender-system library. It automates algorithm selection and hyper parameter optimization in a highly parallelized manner.\nhttps://github.com/ISG-Siegen/Auto-Surprise\n\n**Papers and other**\n\nZero-Shot Recommender Systems\nhttps://arxiv.org/abs/2105.08318\n\nGraph Learning based Recommender Systems\nhttps://arxiv.org/abs/2105.06339\n\nAutoML for Deep Recommender Systems: A Survey\nhttps://arxiv.org/abs/2203.13922\n\nAn Overview of Recommender Systems and Machine Learning in Feature Modeling and Configuration\nhttps://arxiv.org/abs/2102.06634\n\nA Survey of Recommender System Techniques and the Ecommerce Domain\nhttps://arxiv.org/abs/2208.07399\n\nEvalRS: a Rounded Evaluation of Recommender Systems\nhttps://arxiv.org/abs/2207.05772\n\nBehavior Sequence Transformer for E-commerce Recommendation in Alibaba\nhttps://arxiv.org/abs/1905.06874\nhttps://keras.io/examples/structured_data/movielens_recommendations_transformers/\n\nACM Conference on Recommender Systems Years 2007-2022\nWith for example below from RECSYS 2022 (SEATTLE) tutorials and ideas for further reading.\nhttps://recsys.acm.org/recsys22/tutorials/\n- Neural Re-ranking for Multi-stage Recommender Systems.\n- Hands-on Reinforcement learning for recommender systems – From Bandits to SlateQ to Offline RL with Ray RLlib.\n- Offline Evaluation for Group Recommender Systems.\n- Training and Deploying Multi-Stage Recommender Systems.\n- Improving Recommender Systems with Human-in-the-Loop.\n- Hands on Explainable Recommender Systems with Knowledge Graphs.\n- Psychology-informed Recommender Systems.\n- Conversational Recommender System Using Deep Reinforcement Learning.\n" } }