pith. sign in

arxiv: 2410.15272 · v3 · submitted 2024-10-20 · 💻 cs.IR · cs.AI

Performance-Driven QUBO for Recommender Systems on Quantum Annealers

Jiayang Niu , Jie Li , Ke Deng , Mark Sanderson , Nicola Ferro , Yongli Ren This is my paper

Pith reviewed 2026-05-23 19:10 UTC · model grok-4.3

classification 💻 cs.IR cs.AI
keywords QUBOquantum annealersfeature selectionrecommender systemscounterfactual analysisCTR predictionperformance-driven optimization
0
0 comments X

The pith

PDQUBO quantifies the performance impact of features and pairs via counterfactuals to build QUBO problems for quantum annealers in recommender systems.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces PDQUBO, a QUBO-based feature selection method for recommender systems that runs directly on quantum annealers. It uses counterfactual analysis to measure how much each feature and each feature pair contributes to actual model performance, so the optimization objective matches recommendation quality. The formulation stays independent of any specific recommender architecture or evaluation metric. Experiments on real-world datasets show PDQUBO beats earlier QUBO methods on quantum hardware and competes with classical feature selection on CTR prediction tasks. The work also tracks how quantum annealing stability changes with problem size and difficulty.

Core claim

PDQUBO is a performance-driven QUBO formulation for feature selection that explicitly quantifies the performance impact of both individual features and feature pairs on recommender system models through counterfactual analysis. This alignment between the QUBO objective and model performance ensures the solution direction is tied to recommendation quality. The method is model-agnostic and evaluation-metric-independent, allowing execution on quantum annealers and broad applicability across recommender architectures and assessment criteria.

What carries the argument

Counterfactual analysis to estimate performance contributions of individual features and feature pairs, which populates the linear and quadratic terms of the QUBO objective.

If this is right

  • Solutions produced by PDQUBO on quantum annealers yield feature sets that improve recommender performance.
  • The same method applies without modification to any recommender architecture and any evaluation metric.
  • Quantum annealers can serve as practical hardware for feature selection in production recommender pipelines.
  • PDQUBO maintains performance advantages over prior QUBO approaches even as problem size and difficulty increase.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The counterfactual construction could be tested on non-recommender tasks such as general supervised feature selection if the bias-free assumption holds.
  • Reducing observed quantum instability for larger QUBO instances would directly widen the range of recommender datasets that fit on current annealers.
  • Hybrid pipelines that seed classical solvers with PDQUBO-selected subsets might cut overall feature-selection time for very large catalogs.

Load-bearing premise

Counterfactual analysis can reliably estimate the true performance contribution of each feature and pair across arbitrary recommender models and metrics without introducing its own bias.

What would settle it

Retraining multiple recommender models on the features selected by PDQUBO and checking whether the observed performance gains match the counterfactual estimates used to build the QUBO.

Figures

Figures reproduced from arXiv: 2410.15272 by Jiayang Niu, Jie Li, Ke Deng, Mark Sanderson, Nicola Ferro, Yongli Ren.

Figure 1
Figure 1. Figure 1: The Overview of PDQUBO 4.1 Counterfactual Instances Counterfactual Analysis adds perturbations to the base model’s input variables and observes the changes before and after the per￾turbations [37, 52, 59, 67]. In this paper, we refer to these changes as counterfactual instances. Specifically, similar to [52, 67], in the context of feature selection for recommender systems, we measure the impact of item fea… view at source ↗
Figure 2
Figure 2. Figure 2: Energy 𝑌 vs nDCG@10 In this section, we attempt to answer why and how the proposed PDQUBO drives the optimization direction of the QUBO problem toward the optimization of recommendation performance. First, the aim of QUBO is to minimize the energy value 𝑌 as defined in Equation 1. So, it is reasonable to assume: that because PDQUBO is performance-driven, there must be a clear correlation between the minimi… view at source ↗
Figure 3
Figure 3. Figure 3: (a) Distribution of Energy values after selecting 90% of the features using QA. (b) This figure shows the lowest energy values achieved [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
read the original abstract

Quantum annealers offer a promising hardware platform for solving combinatorial optimization problems, especially those formulated as Quadratic Unconstrained Binary Optimization (QUBO). In this work, we propose PDQUBO (Performance-Driven Quadratic Unconstrained Binary Optimization), a QUBO-based feature selection method that is directly executable on quantum annealers. Unlike prior QUBO-based feature selection approaches on quantum annealers, PDQUBO explicitly quantifies the performance impact of both individual features and feature pairs on recommender system models. This alignment between QUBO optimization objectives and model performance ensures that the solution direction is closely tied to recommendation quality, making it well-suited for practical deployment on quantum hardware. Moreover, by leveraging counterfactual analysis, PDQUBO is model-agnostic and evaluation-metric-independent, making it broadly applicable across diverse recommender architectures and assessment criteria. In addition, we investigate the instability of quantum annealing on real quantum devices with respect to varying problem sizes and problem difficulties. Extensive experiments on real-world datasets demonstrate that PDQUBO consistently outperforms prior QUBO-based feature selection methods on quantum annealers. Furthermore, we compare PDQUBO against classical feature selection baselines on click-through rate (CTR) prediction tasks, showing its strong performance and highlighting the potential of using quantum annealers for real-world feature selection applications. Our findings suggest that integrating quantum optimization with counterfactual analysis provides a promising direction for effective feature selection in recommender systems.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The paper proposes PDQUBO, a QUBO-based feature selection method for recommender systems executable on quantum annealers. It quantifies the performance impact of individual features and feature pairs via counterfactual analysis to align the QUBO objective directly with recommendation quality, claiming the resulting formulation is model-agnostic and evaluation-metric-independent. Experiments on real-world datasets show PDQUBO outperforming prior QUBO feature selection methods on quantum hardware and classical baselines on CTR prediction tasks; the work also examines instability of quantum annealing with problem size and difficulty.

Significance. If the counterfactual step produces unbiased, model-independent performance deltas that can be inserted into the QUBO matrix without hidden architecture-specific adjustments, the approach would provide a concrete way to couple quantum optimization objectives with downstream recommender metrics, which is a notable strength for practical quantum-ML applications. The explicit treatment of feature-pair interactions and the reported outperformance on real datasets would further strengthen the case for quantum annealers in feature selection.

major comments (1)
  1. [Abstract] Abstract: the central claim that PDQUBO is 'model-agnostic and evaluation-metric-independent' rests on counterfactual analysis delivering reliable, unbiased performance deltas for features and pairs. The abstract supplies no description of the counterfactual procedure (e.g., how deltas are estimated, whether retraining or surrogates are used, or how correlated features are handled), leaving the model-agnostic assertion unverified and directly exposed to the selection-bias concern raised in the stress-test note. This is load-bearing for the novelty and applicability statements.
minor comments (1)
  1. The abstract refers to 'extensive experiments on real-world datasets' and 'strong performance' but supplies no dataset names, sizes, metrics, or error bars; these details should be added to the abstract or a summary table for immediate assessment of the claimed gains.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback on the abstract. We address the comment below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that PDQUBO is 'model-agnostic and evaluation-metric-independent' rests on counterfactual analysis delivering reliable, unbiased performance deltas for features and pairs. The abstract supplies no description of the counterfactual procedure (e.g., how deltas are estimated, whether retraining or surrogates are used, or how correlated features are handled), leaving the model-agnostic assertion unverified and directly exposed to the selection-bias concern raised in the stress-test note. This is load-bearing for the novelty and applicability statements.

    Authors: We agree the abstract is too concise and does not describe the counterfactual procedure, which weakens the model-agnostic claim for readers. The body (Section 3) specifies that deltas are obtained by measuring the change in a chosen performance metric when a feature or pair is toggled on a fixed validation set using the target recommender's scoring function; no full retraining occurs per subset because incremental forward passes suffice. Pairwise QUBO terms directly encode joint performance effects, which addresses correlation-induced bias. To make the abstract self-contained we will add one sentence summarizing this estimation approach. The stress-test note is not reproduced in the report we received; if it identifies a specific bias not mitigated by the pairwise terms or validation-set evaluation, we request the details so we can respond or add discussion. revision: yes

Circularity Check

0 steps flagged

No circularity detected; derivation relies on external counterfactual analysis without self-referential reduction

full rationale

The provided abstract and summary describe PDQUBO as constructing a QUBO matrix from performance impacts estimated via counterfactual analysis, claiming model-agnostic and metric-independent properties. No equations, derivation steps, or self-citations are visible in the given text. The central construction step (counterfactual deltas inserted into QUBO) is presented as an application of a standard external technique rather than a self-definition, fitted input renamed as prediction, or load-bearing self-citation chain. Without explicit mathematical reductions in the manuscript excerpt, no load-bearing step reduces to its inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review reveals no explicit free parameters, axioms, or invented entities; the method builds on standard QUBO and counterfactual analysis without additional postulates visible here.

pith-pipeline@v0.9.0 · 5799 in / 1122 out tokens · 25721 ms · 2026-05-23T19:10:22.671362+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

77 extracted references · 77 canonical work pages · 3 internal anchors

  1. [1]

    Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions,

    G. Adomavicius and A. Tuzhilin. 2005. Toward the next generation of rec- ommender systems: a survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering 17, 6 (2005), 734–749. https://doi.org/10.1109/TKDE.2005.99

    work page doi:10.1109/tkde.2005.99 2005

  2. [2]

    Akshay Ajagekar and Fengqi You. 2023. Molecular design with automated quantum computing-based deep learning and optimization. Npj Computational Materials 9, 1 (2023), 143

    work page 2023

  3. [3]

    Tameem Albash and Daniel A Lidar. 2018. Demonstration of a scaling advantage for a quantum annealer over simulated annealing. Physical Review X 8, 3 (2018), 031016

    work page 2018

  4. [4]

    Alexandre Belloni, Tengyuan Liang, Hariharan Narayanan, and Alexander Rakhlin. 2015. Escaping the Local Minima via Simulated Annealing: Opti- mization of Approximately Convex Functions. CoRR abs/1501.07242 (2015). arXiv:1501.07242 http://arxiv.org/abs/1501.07242

    work page arXiv 2015

  5. [5]

    Dimitris Bertsimas and John Tsitsiklis. 1993. Simulated annealing. Statistical science 8, 1 (1993), 10–15

    work page 1993

  6. [6]

    Munish Bhatia and Sandeep K Sood. 2020. Quantum computing-inspired network optimization for IoT applications. IEEE Internet of Things Journal 7, 6 (2020), 5590–5598

    work page 2020

  7. [7]

    Sergio Boixo, Vadim N Smelyanskiy, Alireza Shabani, Sergei V Isakov, Mark Dykman, Vasil S Denchev, Mohammad H Amin, Anatoly Yu Smirnov, Masoud Mohseni, and Hartmut Neven. 2016. Computational multiqubit tunnelling in programmable quantum annealers. Nature communications 7, 1 (2016), 10327

    work page 2016

  8. [8]

    Avinash Chalumuri, Raghavendra Kune, and BS Manoj. 2021. A hybrid classical- quantum approach for multi-class classification. Quantum Information Processing 20, 3 (2021), 119

    work page 2021

  9. [9]

    Wei Chen, Yiqing Wu, Zhao Zhang, Fuzhen Zhuang, Zhongshi He, Ruobing Xie, and Feng Xia. 2024. FairGap: Fairness-aware recommendation via generating counterfactual graph. ACM Transactions on Information Systems 42, 4 (2024), 1–25

    work page 2024

  10. [10]

    Xu Chen, Yongfeng Zhang, and Ji-Rong Wen. 2022. Measuring" why" in rec- ommender systems: A comprehensive survey on the evaluation of explainable recommendation. arXiv preprint arXiv:2202.06466 (2022)

    work page arXiv 2022

  11. [11]

    Yulin Chi, Jieshan Huang, Zhanchuan Zhang, Jun Mao, Zinan Zhou, Xiaojiong Chen, Chonghao Zhai, Jueming Bao, Tianxiang Dai, Huihong Yuan, et al. 2022. A programmable qudit-based quantum processor. Nature communications 13, 1 (2022), 1166

    work page 2022

  12. [12]

    Iris Cong, Soonwon Choi, and Mikhail D Lukin. 2019. Quantum convolutional neural networks. Nature Physics 15, 12 (2019), 1273–1278

    work page 2019

  13. [13]

    Washington Cunha, Celso França, Guilherme Fonseca, Leonardo Rocha, and Marcos André Gonçalves. 2023. An effective, efficient, and scalable confidence- based instance selection framework for transformer-based text classification. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval . 665–674

    work page 2023

  14. [14]

    Prasanna Date, Davis Arthur, and Lauren Pusey-Nazzaro. 2021. QUBO formula- tions for training machine learning models. Scientific reports 11, 1 (2021), 10029

    work page 2021

  15. [15]

    Aminu Da’u and Naomie Salim. 2020. Recommendation system based on deep learning methods: a systematic review and new directions. Artificial Intelligence Review 53, 4 (2020), 2709–2748

    work page 2020

  16. [16]

    Michel H Devoret, Andreas Wallraff, and John M Martinis. 2004. Superconducting qubits: A short review. arXiv preprint cond-mat/0411174 (2004)

    work page internal anchor Pith review Pith/arXiv arXiv 2004

  17. [17]

    Guglielmo Faggioli, Nicola Ferro, Raffaele Perego, and Nicola Tonellotto. 2024. Dimension importance estimation for dense information retrieval. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1318–1328

    work page 2024

  18. [18]

    Maurizio Ferrari Dacrema, Fabio Moroni, Riccardo Nembrini, Nicola Ferro, Guglielmo Faggioli, and Paolo Cremonesi. 2022. Towards feature selection for ranking and classification exploiting quantum annealers. InProceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2814–2824

    work page 2022

  19. [19]

    Artur J Ferreira and Mário AT Figueiredo. 2012. Efficient feature selection filters for high-dimensional data. Pattern recognition letters 33, 13 (2012), 1794–1804

    work page 2012

  20. [20]

    Aaron Fisher, Cynthia Rudin, and Francesca Dominici. 2019. All models are wrong, but many are useful: Learning a variable’s importance by studying an entire class of prediction models simultaneously. Journal of Machine Learning Research 20, 177 (2019), 1–81

    work page 2019

  21. [21]

    Jerome H Friedman. 2001. Greedy function approximation: a gradient boosting machine. Annals of statistics (2001), 1189–1232

    work page 2001

  22. [22]

    Yingqiang Ge, Shuchang Liu, Zuohui Fu, Juntao Tan, Zelong Li, Shuyuan Xu, Yunqi Li, Yikun Xian, and Yongfeng Zhang. 2022. A survey on trustworthy recommender systems. ACM Transactions on Recommender Systems (2022)

    work page 2022

  23. [23]

    Fatih Gedikli, Dietmar Jannach, and Mouzhi Ge. 2014. How should I explain? A comparison of different explanation types for recommender systems. Interna- tional Journal of Human-Computer Studies 72, 4 (2014), 367–382

    work page 2014

  24. [24]

    Fred Glover. 1990. Tabu search: A tutorial. Interfaces 20, 4 (1990), 74–94

    work page 1990

  25. [25]

    Fred Glover, Gary Kochenberger, and Yu Du. 2018. A tutorial on formulating and using QUBO models. arXiv preprint arXiv:1811.11538 (2018)

    work page arXiv 2018

  26. [26]

    Huifeng Guo, Ruiming Tang, Yunming Ye, Zhenguo Li, and Xiuqiang He. 2017. DeepFM: a factorization-machine based neural network for CTR prediction.arXiv preprint arXiv:1703.04247 (2017)

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  27. [27]

    Yi Guo, Zhaocheng Liu, Jianchao Tan, Chao Liao, Sen Yang, Lei Yuan, Dongying Kong, Zhi Chen, and Ji Liu. 2022. LPFS: Learnable Polarizing Feature Selection for Click-Through Rate Prediction. arXiv preprint arXiv:2206.00267 (2022)

    work page arXiv 2022

  28. [28]

    Xiangnan He, Kuan Deng, Xiang Wang, Yan Li, Yongdong Zhang, and Meng Wang. 2020. Lightgcn: Simplifying and powering graph convolution network for recommendation. In Proceedings of the 43rd International ACM SIGIR conference on research and development in Information Retrieval . 639–648

    work page 2020

  29. [29]

    Xiangnan He, Lizi Liao, Hanwang Zhang, Liqiang Nie, Xia Hu, and Tat-Seng Chua. 2017. Neural collaborative filtering. In Proceedings of the 26th international conference on world wide web . 173–182

    work page 2017

  30. [30]

    Jonathan L Herlocker, Joseph A Konstan, Loren G Terveen, and John T Riedl

    work page

  31. [31]

    ACM Transactions on Information Systems (TOIS) 22, 1 (2004), 5–53

    Evaluating collaborative filtering recommender systems. ACM Transactions on Information Systems (TOIS) 22, 1 (2004), 5–53

    work page 2004

  32. [32]

    Tongwen Huang, Zhiqi Zhang, and Junlin Zhang. 2019. FiBiNET: combining fea- ture importance and bilinear feature interaction for click-through rate prediction. In Proceedings of the 13th ACM conference on recommender systems . 169–177

    work page 2019

  33. [33]

    Kalervo Järvelin and Jaana Kekäläinen. 2002. Cumulated gain-based evaluation of IR techniques. ACM Transactions on Information Systems (TOIS) 20, 4 (2002), 422–446

    work page 2002

  34. [34]

    Pengyue Jia, Yejing Wang, Zhaocheng Du, Xiangyu Zhao, Yichao Wang, Bo Chen, Wanyu Wang, Huifeng Guo, and Ruiming Tang. 2024. Erase: Benchmarking feature selection methods for deep recommender systems. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 5194– 5205

    work page 2024

  35. [35]

    Tadashi Kadowaki and Hidetoshi Nishimori. 1998. Quantum annealing in the transverse Ising model. Phys. Rev. E 58 (Nov 1998), 5355–5363. Issue 5. https: //doi.org/10.1103/PhysRevE.58.5355

    work page doi:10.1103/physreve.58.5355 1998

  36. [36]

    Thaddeus D Ladd, Fedor Jelezko, Raymond Laflamme, Yasunobu Nakamura, Christopher Monroe, and Jeremy Lloyd O’Brien. 2010. Quantum computers. nature 464, 7285 (2010), 45–53

    work page 2010

  37. [37]

    Weilin Lin, Xiangyu Zhao, Yejing Wang, Tong Xu, and Xian Wu. 2022. AdaFS: Adaptive feature selection in deep recommender system. InProceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 3309–3317

    work page 2022

  38. [38]

    Yaxu Liu, Jui-Nan Yen, Bowen Yuan, Rundong Shi, Peng Yan, and Chih-Jen Lin

    work page

  39. [39]

    In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

    Practical counterfactual policy learning for top-k recommendations. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 1141–1151

    work page

  40. [40]

    Seth Lloyd, Masoud Mohseni, and Patrick Rebentrost. 2014. Quantum principal component analysis. Nature physics 10, 9 (2014), 631–633

    work page 2014

  41. [41]

    Rafael G Mantovani, André LD Rossi, Joaquin Vanschoren, Bernd Bischl, and André CPLF De Carvalho. 2015. Effectiveness of random search in SVM hyper- parameter tuning. In 2015 international joint conference on neural networks (IJCNN). Ieee, 1–8

    work page 2015

  42. [42]

    McGeoch and Pau Farr´e

    Catherine C. McGeoch and Pau Farr´e. 2020. The D-Wave Advantage System: An Overview TECHNICAL REPORT. https://api.semanticscholar.org/CorpusID: 222355754

    work page 2020

  43. [43]

    Satoshi Morita and Hidetoshi Nishimori. 2008. Mathematical foundation of quantum annealing. J. Math. Phys. 49, 12 (2008)

    work page 2008

  44. [44]

    Sascha Mücke, Raoul Heese, Sabine Müller, Moritz Wolter, and Nico Piatkowski

    work page

  45. [45]

    Quantum Machine Intelligence 5, 1 (2023), 11

    Feature selection on quantum computers. Quantum Machine Intelligence 5, 1 (2023), 11

    work page 2023

  46. [46]

    Riccardo Nembrini, Maurizio Ferrari Dacrema, and Paolo Cremonesi. 2021. Fea- ture selection for recommender systems with quantum computing. Entropy 23, 8 (2021), 970

    work page 2021

  47. [47]

    Andrea Pasin, Washington Cunha, Marcos André Gonçalves, and Nicola Ferro

    work page

  48. [48]

    In Proceedings of the 2024 ACM SIGIR International Conference on Theory of Information Retrieval

    A quantum annealing instance selection approach for efficient and effective transformer fine-tuning. In Proceedings of the 2024 ACM SIGIR International Conference on Theory of Information Retrieval . 205–214

    work page 2024

  49. [49]

    Davide Pastorello and Enrico Blanzieri. 2019. Quantum annealing learning search for solving QUBO problems. Quantum Information Processing 18, 10 (2019), 303

    work page 2019

  50. [50]

    Hanchuan Peng, Fuhui Long, and Chris Ding. 2005. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min- redundancy. IEEE Transactions on pattern analysis and machine intelligence 27, 8 (2005), 1226–1238

    work page 2005

  51. [51]

    Atanu Rajak, Sei Suzuki, Amit Dutta, and Bikas K Chakrabarti. 2023. Quantum annealing: An overview. Philosophical Transactions of the Royal Society A 381, 2241 (2023), 20210417

    work page 2023

  52. [52]

    Patrick Rebentrost, Masoud Mohseni, and Seth Lloyd. 2014. Quantum support vector machine for big data classification. Physical review letters 113, 13 (2014), 130503

    work page 2014

  53. [53]

    Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt-Thieme

    work page

  54. [54]

    BPR: Bayesian Personalized Ranking from Implicit Feedback

    BPR: Bayesian personalized ranking from implicit feedback. arXiv preprint arXiv:1205.2618 (2012). Performance-Driven QUBO for Recommender Systems on Quantum Annealers Conference acronym ’XX, June 03–05, 2018, Woodstock, NY

    work page internal anchor Pith review Pith/arXiv arXiv 2012

  55. [55]

    Badrul Sarwar, George Karypis, Joseph Konstan, and John Riedl. 2001. Item-based collaborative filtering recommendation algorithms. In Proceedings of the 10th international conference on World Wide Web. 285–295

    work page 2001

  56. [56]

    Juntao Tan, Shijie Geng, Zuohui Fu, Yingqiang Ge, Shuyuan Xu, Yunqi Li, and Yongfeng Zhang. 2022. Learning and evaluating graph neural network explana- tions based on counterfactual and factual reasoning. In Proceedings of the ACM web conference 2022. 1018–1027

    work page 2022

  57. [57]

    Juntao Tan, Shuyuan Xu, Yingqiang Ge, Yunqi Li, Xu Chen, and Yongfeng Zhang

    work page

  58. [58]

    InProceedings of the 30th ACM International Conference on Information & Knowledge Management

    Counterfactual explainable recommendation. InProceedings of the 30th ACM International Conference on Information & Knowledge Management . 1784–1793

    work page

  59. [59]

    Robert Tibshirani. 1996. Regression shrinkage and selection via the lasso.Journal of the Royal Statistical Society Series B: Statistical Methodology 58, 1 (1996), 267– 288

    work page 1996

  60. [60]

    Nava Tintarev and Judith Masthoff. 2015. Explaining recommendations: Design and evaluation. In Recommender systems handbook. Springer, 353–382

    work page 2015

  61. [61]

    Kari Torkkola. 2003. Feature extraction by non-parametric mutual information maximization. Journal of machine learning research 3, Mar (2003), 1415–1438

    work page 2003

  62. [62]

    Khanh Hiep Tran, Azin Ghazimatin, and Rishiraj Saha Roy. 2021. Counterfactual explanations for neural recommenders. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . 1627–1631

    work page 2021

  63. [63]

    Sahil Verma, Varich Boonsanong, Minh Hoang, Keegan Hines, John Dickerson, and Chirag Shah. 2020. Counterfactual explanations and algorithmic recourses for machine learning: A review. Comput. Surveys (2020)

    work page 2020

  64. [64]

    Jianwei Wang, Stefano Paesani, Raffaele Santagati, Sebastian Knauer, Antonio A Gentile, Nathan Wiebe, Maurangelo Petruzzella, Jeremy L O’brien, John G Rarity, Anthony Laing, et al. 2017. Experimental quantum Hamiltonian learning. Nature Physics 13, 6 (2017), 551–555

    work page 2017

  65. [65]

    Xiangmeng Wang, Qian Li, Dianer Yu, Qing Li, and Guandong Xu. 2024. Coun- terfactual explanation for fairness in recommendation. ACM Transactions on Information Systems 42, 4 (2024), 1–30

    work page 2024

  66. [66]

    Xiangmeng Wang, Qian Li, Dianer Yu, Qing Li, and Guandong Xu. 2024. Re- inforced path reasoning for counterfactual explainable recommendation. IEEE Transactions on Knowledge and Data Engineering (2024)

    work page 2024

  67. [67]

    Yejing Wang, Zhaocheng Du, Xiangyu Zhao, Bo Chen, Huifeng Guo, Ruiming Tang, and Zhenhua Dong. 2023. Single-shot feature selection for multi-task recommendations. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval . 341–351

    work page 2023

  68. [68]

    Yejing Wang, Xiangyu Zhao, Tong Xu, and Xian Wu. 2022. Autofield: Automating feature selection in deep recommender systems. In Proceedings of the ACM Web Conference 2022. 1977–1986

    work page 2022

  69. [69]

    Zhenlei Wang, Jingsen Zhang, Hongteng Xu, Xu Chen, Yongfeng Zhang, Wayne Xin Zhao, and Ji-Rong Wen. 2021. Counterfactual data-augmented se- quential recommendation. In Proceedings of the 44th international ACM SIGIR conference on research and development in information retrieval . 347–356

    work page 2021

  70. [70]

    Sheir Yarkoni, Elena Raponi, Thomas Bäck, and Sebastian Schmitt. 2022. Quantum annealing for industry applications: Introduction and review. Reports on Progress in Physics 85, 10 (2022), 104001

    work page 2022

  71. [71]

    Dianlong You, Shina Niu, Siqi Dong, Huigui Yan, Zhen Chen, Di Wu, Limin Shen, and Xindong Wu. 2023. Counterfactual explanation generation with minimal feature boundary. Information Sciences 625 (2023), 342–366

    work page 2023

  72. [72]

    Beichuan Zhang, Chenggen Sun, Jianchao Tan, Xinjun Cai, Jun Zhao, Mengqi Miao, Kang Yin, Chengru Song, Na Mou, and Yang Song. 2023. SHARK: A Light- weight Model Compression Approach for Large-scale Recommender Systems. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. 4930–4937

    work page 2023

  73. [73]

    Shengyu Zhang, Dong Yao, Zhou Zhao, Tat-Seng Chua, and Fei Wu. 2021. Causerec: Counterfactual user sequence synthesis for sequential recommen- dation. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval . 367–377

    work page 2021

  74. [74]

    Yongfeng Zhang, Guokun Lai, Min Zhang, Yi Zhang, Yiqun Liu, and Shaoping Ma. 2014. Explicit factor models for explainable recommendation based on phrase-level sentiment analysis. In Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval . 83–92

    work page 2014

  75. [75]

    Linghua Zhu, Ho Lun Tang, George S Barron, FA Calderon-Vargas, Nicholas J Mayhall, Edwin Barnes, and Sophia E Economou. 2022. Adaptive quantum approximate optimization algorithm for solving combinatorial problems on a quantum computer. Physical Review Research 4, 3 (2022), 033029

    work page 2022

  76. [76]

    Yaochen Zhu, Jing Ma, Liang Wu, Qi Guo, Liangjie Hong, and Jundong Li. 2023. Path-specific counterfactual fairness for recommender systems. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 3638–3649

    work page 2023

  77. [77]

    Martin Zinkevich, Markus Weimer, Lihong Li, and Alex Smola. 2010. Parallelized stochastic gradient descent. Advances in neural information processing systems 23 (2010)

    work page 2010