Zheng Wen - Publications

   

Journal Papers

  • [J11] Y. Chen, Z. Wen, and Y. Xie, "Dynamic Pricing in an Evolving and Unknown Marketplace", accepted by Management Science. [SSRN]

Book Chapters

Papers In Progress

  • W. Xu, S. Dong, X. Lu, G. Lam, Z. Wen, and B. Van Roy, "RLHF and IIA: Perverse Incentives", to be submitted. [arXiv]

  • D. Tang, R. Jain, B. Hao, and Z. Wen, "Efficient Online Learning with Offline Datasets for Infinite Horizon MDPs: A Bayesian Approach", to be submitted. [arXiv]

  • Z. Wen, I. Osband, C. Qin, X. Lu, M. Ibrahimi, V. Dwaracherla, S. M. Asghari, and B. Van Roy, "From Predictions to Decisions: The Importance of Joint Predictive Distributions", to be submitted. [arXiv]

  • "The Target Return Strategy", with Prof. Ying Ian Xue and Prof. Xu Jiang, submitted to The Financial Review, third-round revision.

  • P. Grigas, A. Lobos, Z. Wen, and K. Lee, "Optimal Bidding, Allocation, and Budget Spending for a Demand-Side Platform with Generic Auctions", submitted to Production and Operations Management, first-round major revision. [SSRN]

Full Conference Papers

  • [C48] I. Osband, Z. Wen, S. M. Asghari, V. Dwaracherla, M. Ibrahimi, X. Lu, and B. Van Roy, "Epistemic Neural Networks", accepted by the 37th Conference on Neural Information Processing Systems (NeurIPS 2023) as a spotlight. [arXiv]

  • [C38] T. Yu, B. Kveton, Z. Wen, R. Zhang, and O. J. Mengshoel, "Influence Diagram Bandits", Thirty-seventh International Conference on Machine Learning (ICML 2020), online conference, originally planned to be held at Vienna, Austria.

  • [C36] P. Perrault, J. Healey, Z. Wen, and M. Valko, "Budgeted Online Influence Maximization", Thirty-seventh International Conference on Machine Learning (ICML 2020), online conference, originally planned to be held at Vienna, Austria.

  • [C33] V. Dwaracherla, X. Lu, M. Ibrahimi, I. Osband, Z. Wen, and B. Van Roy, "Hypermodels for Exploration", the Eighth International Conference on Learning Representations (ICLR 2020), online conference, originally planned to be held at Addis Ababa, Ethiopia.

  • [C30] Y. Xue, Z. Wen, and X. Jiang, "The Optimal Reservation Price", accepted for presentation at
    • the 2020 Southern Finance Association (SFA) Annual Meeting, Online Conference.
    • the 59th Annual Southwestern Finance Association (SWFA) Conference, San Antonio, Texas, 2020.
    • the 2019 Southern Finance Association (SFA) Annual Meeting, Orlando, Florida, with title "The Optimal Price Trigger".
  • [C16] S. Katariya, B. Kveton, C. Szepesvari, C. Vernade, and Z. Wen, "Stochastic Rank-1 Bandits", the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), Fort Lauderdale, Florida, 2017. [arXiv]

  • [C3] Z. Wen, B. Kveton, B. Eriksson, and S. Bhamidipati, "Sequential Bayesian Search", International Conference on Machine Learning (ICML), Atlanta, Georgia, 2013. (acceptance rate: 24%)

Short Conference Papers

Workshop Papers

  • [W9] R. Zhang, C. Chen, Z. Gan, Z. Wen, W. Wang, and L. Carin, "Nested-Wasserstein Self-Imitation Learning for Sequence Generation", Deep Reinforcement Learning Workshop, NeurIPS 2019, Vancouver, Canada.

  • [W8] R. Zhang, C. Chen, Z. Gan, W. Wang, Z. Wen, and L. Carin, "Sequence Generation with a Guider Network", ICML'19 Workshop on Real-World Sequential Decision Making, Long Beach, CA.

  • [W7] R. Zhang, Z. Wen, C. Chen, and L. Carin, "Scalable Thompson Sampling via Optimal Transport", Infer to Control Workshop on Probabilistic Reinforcement Learning and Structured Control at NIPS 2018, Montréal, Canada.

  • [W6] A. Lobos, P. Grigas, Z. Wen, and K. Lee, "Optimal Bidding, Allocation and Budget Spending for a Demand Side Platform Under Many Auction Types", AdKDDTargetAd2018 Workshop at KDD 2018, London, United Kingdom. [arXiv] (Best Student Paper)

  • [W5] S. Li, Y. Abbasi-Yadkori, B. Kveton, S. Muthukrishnan, V. Vinay, and Z. Wen, "Offline Evaluation of Ranking Policies with Click Models", CausalML 2018, Stockholm, Sweden.

  • [W4] C. Vernade, B. Kveton, Y. Abbasi-Yadkori, M. Ghavamzadeh, and Z. Wen, "Rank-1 A/B Testing", Women in Machine Learning Workshop 2017.

  • [W3] P. Grigas, A. Lobos, Z. Wen, and K. Lee, "Profit Maximization for Online Advertising Demand-Side Platforms", AdKDDTargetAd2017 Workshop at KDD 2017, Halifax, Canada. [arXiv]

arXiv Preprints & Other Publication

  • X. Lu, I. Osband, S. M. Asghari, S. Gowal, V. Dwaracherla, Z. Wen, B. Van Roy, "Robustness of Epinets against Distributional Shifts", arXiv preprint. [arXiv]

  • W. Mou, Z. Wen, and X. Chen, "On the Sample Complexity of Reinforcement Learning with Policy Space Generalization", arXiv preprint. [arXiv]

  • B. Kveton, S. Mahdian, S. Muthukrishnan, Z. Wen, and Y. Xian, "Waterfall Bandits: Learning to Sell Ads Online", arXiv preprint. [arXiv]

  • S. Vaswani, B. Kveton, Z. Wen, A. Rao, M. Schmidt, and Y. Abbasi-Yadkori, "New Insights into Bootstrapping for Bandits", arXiv preprint. [arXiv]

  • B. Kveton, C. Szepesvari, A. Rao, Z. Wen, Y. Abbasi-Yadkori, and S. Muthukrishnan, "Stochastic Low-Rank Bandits", arXiv preprint. [arXiv]

  • Z. Wen, E. Bax, and J. Li, "Revenue-Maximizing Mechanism Design for Quasi-Proportional Auctions", arXiv preprint. [arXiv]

  • B. Kveton, Z. Wen, A. Ashkan, and M. Valko, "Learning to Act Greedily: Polymatroid Semi-Bandits", arXiv preprint. [arXiv]

Code

  • Epistemic Neural Networks. This is a general interface for uncertainty modeling in deep learning. All existing approaches to uncertainty modeling, such as Bayesian neural networks (BNNs), can be expressed as ENNs. However, there are ENN architectures that can not be expressed as BNNs. This library provides interfaces and tools for designing and training ENNs.

  • The Neural Testbed. This is a system for evaluating the performance of epistemic neural networks, which are models that generate joint predictions over multiple outputs in response to multiple inputs. We have also included a set of baseline agents.

Patents & Filed Patents

  • [P10] Efficient Exploration of Offline Models to Warm Start Online Bandit Learning, with G. Theocharous, Y. Abbasi-Yadkori, and Q. Wu, US Patent App., filed, 2019.

  • [P9] Multi-Task Equidistant Embedding, with H. Zhao, S. Kim, S. Li and B. Kveton, US Patent App., filed, 2018.

  • [P8] Change Point Detection in a Multi-Armed Bandit Recommendation System, with Y. Cao and B. Kveton, US Patent App., filed, 2018.

  • [P7] Online Training of Segmentation Model via Interactions With Interactive Computing Environment, with T. Yu, B. Kveton, and H. Bui, US Patent App., filed, 2018.

  • [P6] Multivariate Digital Campaign Content Testing Utilizing Rank-1 Best-Arm Identification, with Y. Abbasi-Yadkori, M. Ghavamzadeh, B. Kveton, and C. Vernade, US Patent App., filed, 2018.

  • [P5] Training And Utilizing Item-level Importance Sampling Models for Offline Evaluation and Execution of Digital Content Selection Policies, with S. Li, Y. Abbasi-Yadkori, B. Kveton, and V. Vinay, US Patent App., filed, 2018.

  • [P4] Online Diverse Set Generation From Partial-Click Feedback, with B. Kveton, P. Gupta, I. A. Burhanuddin, H. Singh, and G. Hiranandani, US Patent App., filed, 2018.

  • [P3] Influence Maximization Determination in a Social Network System, with S. Vaswani, B. Kveton, and M. Ghavamzadeh, US Patent App., filed, 2017.

Theses