/RecDebiasing

This repository collects debiasing methods for recommendation

This website collects recent works and datasets on recommendation debiasing and their codes. We hope this website could help you do search on this topic.

We will keep updating this list, and if you find any missing related work or have any suggestions, please feel free to contact us (cjwustc@ustc.edu.cn). Thank you!

Relevant Papers/Codes on Recommendation Debiasing

More works about Introduction of bias can be found in this file:

Debiasing Strategies

Papers bias Methods Conference Code
Causal Embeddings for Recommendation Universal CausE Recsys 2018 Tensorflow
A General Knowledge Distillation Framework for Counterfactual Recommendation via Uniform Data Universal KDCRec SIGIR 2020 Tensorflow
AutoDebias: Learning to Debias for Recommendation Universal AutoDebias SIGIR 2021 Pytorch
Transfer Learning in Collaborative Recommendation for Bias Reduction Universal TJR Recsys 2021 Tensorflow
Collaborative filtering and the missing at random assumption Selection Bias UAI 2007 Python
Collaborative prediction and ranking with non-random missing data Selection Bias MM RecSys 2009
Training and testing of recommender systems on data missing not at random Selection Bias KDD 2010
Training and testing of recommender systems on data missing not at random Selection Bias KDD 2010
Evaluation of recommendations: rating-prediction and ranking Selection Bias RecSys 2013
Probabilistic matrix factorization with non-random missing data Selection Bias MF-MNAR JMLR 2014 Python
Bayesian Binomial Mixture Model for Collaborative Prediction With Non-Random Missing Data Selection Bias RecSys 2014
Boosting Response Aware Model-Based Collaborative Filtering Selection Bias RAPMF TKDE 2015 -
Recommendations as treatments: Debiasing learning and evaluation Selection Bias MF-IPS PMLR 2016 Python
Social recommendation with missing not at random data Selection Bias SPMF-MNAR ICDM 2018
The deconfounded recommender: A causal inference approach to recommendation Selection Bias Deconfounded model arXiv 2018
Doubly robust joint learning for recommendation on data missing not at random Selection Bias DR ICML 2019
Recommendations as treatments: Debiasing learning and evaluation Selection Bias MF-IPS SIGIR 2020 Python
Asymmetric tri-training for debiasing missing-not-at-random explicit feedback Selection Bias AT SIGIR 2020
Combating Selection Biases in Recommender Systems with a Few Unbiased Ratings Selection Bias LTD WSDM 2021
Learning to recommend with social trust ensemble Conformity Bias RSTE SIGIR 2009
mtrust: discerning multi-faceted trust in a connected world Conformity Bias mTrust WSDM 2012
A probabilistic model for using social networks in personalized item recommendation Conformity Bias SPF RecSys 2015 Python
Are you influenced by others when rating?: Improve rating prediction by conformity modeling Conformity Bias RecSys 2016
Xgboost: A scalable tree boosting system Conformity Bias XGBoost KDD 2016 Python
Learning personalized preference of strong and weak ties for social recommendation Conformity Bias PTPMF WWW 2017
Disentangling user interest and Conformity for recommendation with causal embedding Conformity Bias DICE WWW 2021 Pytorch
Popularity Bias Is Not Always Evil: Disentangling Benign and Harmful Bias for Recommendation Conformity Bias TIDE Arxiv 2021 To publish
Collaborative filtering for implicit feedback datasets Exposure Bias ICDM 2008 Python
One-class collaborative filtering Exposure Bias wALS/sALS ICDM 2008
Mind the gaps: weighting the unknown in large-scale one-class collaborative filtering Exposure Bias GALS/sMMMF KDD 2009
Bpr: Bayesian personalized ranking from implicit feedback Exposure Bias BPR UAI 2009 Python
Improving one-class collaborative filtering by incorporating rich user information Exposure Bias CIKM 2010
Logistic matrix factorization for implicit feedback data Exposure Bias LogisticMF NIPS 2014
Dynamic matrix factorization with priors on unknown values Exposure Bias KDD 2015 C++
Fast matrix factorization for online recommendation with implicit feedback Exposure Bias eALS SIGIR 2016 Python
Collaborative denoising auto-encoders for top-n recommender systems （CDAE） Exposure Bias CDAE WSDM 2016 C++
Modeling user exposure in recommendation Exposure Bias Content ExpoMF/Location ExpoMF WWW 2016 Python
Learning to rank with selection bias in personal search Exposure Bias Global/Segmented/ Generalized Bias Model SIGIR 2016
Selection of negative samples for one-class matrix factorization Exposure Bias Full SDM 2017 Python
Neural collaborative filtering （NCF） Exposure Bias NCF WWW 2017 Python
Selection of negative samples for one-class matrix factorization Exposure Bias Full SDM 2017 Python
Unbiased offline recommender evaluation for missing-not-at-random implicit feedback Exposure Bias RecSys 2018 Python
An improved sampler for bayesian personalized ranking by leveraging view data Exposure Bias BPR+view WWW 2018
Collaborative filtering with social exposure: A modular approach to social recommendation Exposure Bias SERec AAAI 2018 C++
Entire space multi-task model: An effective approach for estimating post-click conversion rate Exposure Bias ESMM SIGIR 2018
“Modeling users’ exposure with social knowledge influence and consumption influence for recommendation Exposure Bias SoEXBMF CIKM 2018
Reinforced negative sampling for recommendation with exposure data Exposure Bias RNS IJCAI 2019 Python
Samwalker: Social recommendation with informative sampling strategy Exposure Bias Samwalker WWW 2019 MATLAB/C++
Samwalker: Social recommendation with informative sampling strategy Exposure Bias Samwalker WWW 2019 MATLAB/C++
Unbiased recommender learning from missing-not-at-random implicit feedback Exposure Bias Rel-MF WSDM 2020 Python
Lightgcn: Simplifying and powering graph convolution network for recommendation （LightGCN） Exposure Bias LightGCN SIGIR 2020 Python
Reinforced negative sampling over knowledge graph for recommendation Exposure Bias KGPolicy WWW 2020 Python
Fast adaptively weighted matrix factorization for recommendation with implicit feedback Exposure Bias FAWMF AAAI 2020
Correcting for selection bias in learning-to-rank systems Exposure Bias Heckman rank/Propensity SVM rank WWW 2020 Python
Large-scale causal approaches to debiasing post-click conversion rate estimation with multi-task learning Exposure Bias Multi-IPW/Multi-DR WWW 2020
Entire space multi-task modeling via post-click behavior decomposition for conversion rate prediction Exposure Bias ESM^2^ SIGIR 2020
Gmcm: Graph-based micro-behavior conversion model for post-click conversion rate estimation Exposure Bias GMCM SIGIR 2020
” click” is not equal to” like”: Counterfactual recommendation for mitigating clickbait issue Exposure Bias CR arXiv 2020
Unbiased Implicit Recommendation and Propensity Estimation via Combinational Joint Learning Exposure Bias CJ Recsys 2020 Tensorflow
Debiasing Item-to-Item Recommendations With Small Annotated Datasets Exposure Bias - Recsys 2020 -
SamWalker++: recommendation with informative sampling strategy Exposure Bias SamWalker++ TKDE 2021 Pytorch
Top-N Recommendation with Counterfactual User Preference Simulation Exposure Bias CPR CIKM 2021 -
Debiased Explainable Pairwise Ranking from Implicit Feedback Exposure Bias EBPR Recsys 2021 Pytorch
Deconfounded Causal Collaborative Filtering Exposure Bias DCCF Arxiv 2021 -
Mitigating Confounding Bias in Recommendation via Information Bottleneck Exposure Bias DIB Recsys 2021 To come
Propensity-Independent Bias Recovery in Offline Learning-to-Rank Systems Exposure Bias LTR SIGIR 2021 -
Non-Clicks Mean Irrelevant? Propensity Ratio Scoring As a Correction Exposure Bias PRS WSDM 2021 -
Comparing click logs and editorial labels for training query rewriting Position Bias WWW 2007
An experimental comparison of click position-bias models Position Bias WSDM 2008
A user browsing model to predict search engine click data from past observations. Position Bias SIGIR 2008
A dynamic bayesian network click model for web search ranking Position Bias DBN WWW 2009
Click chain model in web search Position Bias CCM WWW 2009
A novel click model and its applications to online advertising Position Bias GCM WSDM 2010
Learning socially optimal information systems from egoistic users Position Bias SoPer-R/SoPer-S ECML PKDD 2013
Reusing historical interaction data for faster online learning to rank for ir Position Bias RHC/CPS WSDM 2013 Link
Batch learning from logged bandit feedback through counterfactual risk minimization Position Bias POEM JMLR 2015
Learning to rank with selection bias in personal search Position Bias Global/Segmented/ Generalized Bias Model SIGIR 2016
Multileave gradient descent for fast online learning to rank Position Bias WSDM 2016
Unbiased learning-to-rank with biased feedback Position Bias Propensity SVM-Rank WSDM 2017 Link
Unbiased learning to rank with unbiased propensity estimation Position Bias DLA SIGIR 2018 Python
Position bias estimation for unbiased learning to rank in personal search Position Bias Regression-based EM WSDM 2018
Addressing Trust Bias for Unbiased Learning-to-Rank Position Bias TrustPBM WWW 2019 -
A deep recurrent survival model for unbiased ranking Position Bias DRSR SIGIR 2020 Python
Attribute-based propensity for unbiased learning in recommender systems: Algorithm and case studies Position Bias KDD 2020
Debiasing grid-based product search in e-commerce Position Bias KDD 2020
Cascade model-based propensity estimation for counterfactual learning to rank Position Bias CM-IPS SIGIR 2020 Python
When Inverse Propensity Scoring does not Work: Affine Corrections for Unbiased Learning to Rank Position Bias Affine estimator CIKM 2020 Tensorflow
A Graph-Enhanced Click Model for Web Search Position Bias GraphCM SIGIR 2021 Pytorch
Adapting Interactional Observation Embedding for Counterfactual Learning to Rank Position Bias IOBM SIGIR 2021 Tensorflow
Efficiency improvement of neutrality-enhanced recommendation Popularity Bias RecSys 2013 Link
Correcting popularity bias by enhancing recommendation neutrality Popularity Bias RecSys 2014
Incorporating diversity in a learning to rank recommender system Popularity Bias FLAIRS 2016
The limits of popularity-based recommendations, and the role of social ties Popularity Bias KDD 2016 C++
Controlling popularity bias in learning-to-rank recommendation Popularity Bias RecSys 2017
An adversarial approach to improve long-tail performance in neural collaborative filtering Popularity Bias CIKM 2018
Unbiased offline recommender evaluation for missing-not-at-random implicit feedback Popularity Bias RecSys 2018 Python
Popularity bias in ranking and recommendation Popularity Bias AIES 2019
ESAM: discriminative domain adaptation with non-displayed items to improve long-tail performance Popularity Bias ESAM SIGIR 2020 Python
Disentangling User Interest and Conformity for Recommendation with Causal Embedding Popularity Bias DICE WWW 2021 Python
The Unfairness of Popularity Bias in Recommendation Popularity Bias VAE-CF SAC 2021 Pytorch
Popularity Bias in Dynamic Recommendation Popularity Bias FPC KDD 2021 Tensorflow
Causal Intervention for Leveraging Popularity Bias in Recommendation Popularity Bias PDA SIGIR 2021 Tensorflow
Deconfounded Recommendation for Alleviating Bias Amplification Popularity Bias DecRS KDD 2021 Pytorch
Popularity Bias Is Not Always Evil: Disentangling Benign and Harmful Bias for Recommendation Popularity Bias TIDE Arxiv 2021 To come
Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system Popularity Bias MACR KDD 2021 Tensorflow
Discrimination-aware data mining Unfairness KDD 2008
Enhancement of the neutrality in recommendation Unfairness RecSys 2012
Learning fair representations Unfairness JMLR 2013
Efficiency improvement of neutrality-enhanced recommendation. Unfairness RecSys 2013 Link
Model-based approaches for independence-enhanced recommendation Unfairness IEEE 2016 Link
Censoring representations with an adversary Unfairness ALFR ICLR 2016
Fa*ir: A fair top-k ranking algorithm Unfairness FA*IR CIKM 2017 Java/Python
Beyond parity: Fairness objectives for collaborative filtering Unfairness NIPS 2017
Considerations on recommendation independence for a find-good-items task Unfairness IERS Recsys 2017
Balanced neighborhoods for fairness-aware collaborative recommendation Unfairness SLIM RecSys 2017
Fairness-aware group recommendation with pareto-efficiency Unfairness RecSys 2017
New fairness metrics for recommendation that embrace differences Unfairness FAT/ML 2017
Counterfactual fairness Unfairness counterfactual fairness arXiv 2017 Python
Controlling popularity bias in learning-to-rank recommendation Unfairness RecSys 2017
Equity of attention: Amortizing individual fairness in rankings Unfairness SIGIR 2018
Fairness of exposure in rankings Unfairness KDD 2018
Fairness-aware tensor-based recommendation Unfairness FATR CIKM 2018 Pyhton
A fairness-aware hybrid recommender system Unfairness PSL FATREC 2018
Fairness in decision-making - the causal explanation formula Unfairness Ctf-DE/Ctf-IE/Ctf-SE AAAI 2018
On discrimination discovery and removal in ranked data using causal graph Unfairness FRank/FDetect KDD 2018
Fair inference on outcomes Unfairness AAAI 2018 C++
Fairness-aware ranking in search & recommendation systems with application to linkedin talent search Unfairness PSL KDD 2019
Designing fair ranking schemes Unfairness SIGMOD 2019
Fairwalk: Towards fair graph embedding Unfairness Fairwalk IJCAI 2019
Fairness in recommendation ranking through pairwise comparisons Unfairness KDD 2019
Counterfactual fairness: Unidentification bound and algorithm. Unfairness IJCAI 2019
“Compositional fairness constraints for graph embeddings Unfairness ICML 2019 Python
Privacy-aware recommendation with private-attribute protection using adversarial learning Unfairness RAP WSDM 2019
Policy Learning for Fairness in Ranking Unfairness FAIR-PG-RANK NIPS 2019 Python
Debayes: a bayesian method for debiasing network embeddings Unfairness DeBayes ICML 2020 Link
Controlling fairness and bias in dynamic learning-to-rank Unfairness FairCo SIGIR 2020 Python
Measuring and Mitigating Item Under-Recommendation Bias in Personalized Ranking Systems Unfairness DPR SIGIR 2020 Tensorflow
Leave No User Behind: Towards Improving the Utility of Recommender Systems for Non-mainstream Users Unfairness NAECF WSDM 2021 Python
User-oriented Fairness in Recommendation Unfairness WWW2021 Python
Policy-Gradient Training of Fair and Unbiased Ranking Functions Unfairness FULTR SIGIR 2021 Python
Towards Long-term Fairness in Recommendation Unfairness FCPO WSDM 2021 Python
Towards Personalized Fairness based on Causal Notion Unfairness SIGIR 2021
Learning Fair Representations for Recommendation: A Graph-based Perspective Unfairness FairGo WWW 2021 PyTorch
Debiasing Career Recommendations with Neural Fair Collaborative Filtering Unfairness NFCF WWW 2021 PyTorch
Computationally Efficient Optimization of Plackett-Luce Ranking Models for Relevance and Fairness Unfairness PL-Rank SIGIR 2021 Tensorflow
A contextual-bandit approach to personalized news article recommendation Loop Effect LinUCB WWW 2010 Python
Interactive collaborative filtering Loop Effect ICF CIKM 2013 Python
Predicting counterfactuals from large historical data and small randomized trials Loop Effect WWW 2016
Deconvolving feedbackloops in recommender systems Loop Effect Deconvolving feedback NIPS 2016
Interactive social recommendation Loop Effect ISR CIKM 2017
Factorization bandits for interactive recommendation. Loop Effect FactorUCB AAAI 2017
Off-policy evaluation for slate recommendation Loop Effect PI NIPS 2017 Python
Causal embeddings for recommendation Loop Effect CausE RecSys 2018 Python
Stabilizing reinforcement learning in dynamic environment with application to online recommendation Loop Effect Robust DQN KDD 2018
Drn: A deep reinforcement learning framework for news recommendation Loop Effect DDQN WWW 2018
Recommendations with negative feedback via pairwise deep reinforcement learning Loop Effect DEERS KDD 2018
Deep reinforcement learning for page-wise recommendations Loop Effect DeepPage RecSys 2018
A reinforcement learning framework for explainable recommendation Loop Effect ICDM 2018
Degenerate feedback loops in recommender systems Loop Effect Oracle AIES 2019
Improving ad click prediction by considering non-displayed events Loop Effect CIKM 2019 Link
Large-scale interactive recommendation with tree-structured policy gradient Loop Effect TPGR AAAI 2019 Python
When people change their mind: Off-policy evaluation in non-stationary recommendation environments Loop Effect WSDM 2019 Python
Deep reinforcement learning for list-wise recommendations Loop Effect LIRD KDD 2019 Python
Top-k off-policy correction for a reinforce recommender system Loop Effect REINFORCE WSDM 2019 Python
Debiasing the human-recommender system feedback loop in collaborative filtering Loop Effect propensity MF WWW 2019
A general knowledge distillation framework for counterfactual recommendation via uniform data Loop Effect KDCRec SIGIR 2020 Python
Influence function for unbiased recommendation Loop Effect IF4URec SIGIR 2020
Jointly learning to recommend and advertise Loop Effect RAM KDD 2020
Counterfactual evaluation of slate recommendations with sequential reward interactions Loop Effect RIPS KDD 2020 Python
Joint policy value learning for recommendation Loop Effect Dual Bandit KDD 2020 Python
AutoDebias: Learning to Debias for Recommendation Loop Effect AutoDebias SIGIR 2021 PyTorch

Dataset

We collect some datasets which include unbiased data and are often used in the research of recommendation debiasing.

Dataset Paper Source
Yahoo! R3 Collaborative Prediction and Ranking with Non-Random Missing Data Data
Coat Recommendations as Treatments: Debiasing Learning and Evaluation Data
KuaiRec KuaiRec: A Fully-observed Dataset for Recommender Systems Data

Tips

If you find this repository useful to your research or work, it is really appreciate to star this repository.