Conference and Journal Papers

2024

  1. Inverse Reinforcement Learning with Multiple Planning Horizons
    Yao, Jiayu, Pan, Weiwei, Doshi-Velez, Finale, and Engelhardt, Barbara E
    Reinforcement Learning Journal 2024
    HTML

2023

  1. Compositional Q-learning for electrolyte repletion with imbalanced patient sub-populations
    Mandyam, Aishwarya, Jones, Andrew,  Yao, Jiayu, Laudanski, Krzysztof, and Engelhardt, Barbara E
    In Machine Learning for Health (ML4H) 2023
    HTML
  2. Performance Bounds for Model and Policy Transfer in Hidden-parameter MDPs
    Fu, Haotian,  Yao, Jiayu, Gottesman, Omer, Doshi-Velez, Finale, and Konidaris, George
    In The Eleventh International Conference on Learning Representations 2023
    HTML

2021

  1. Power Constrained Bandits
    Yao, Jiayu, Brunskill, Emma, Pan, Weiwei, Murphy, Susan, and Doshi-Velez, Finale
    In Machine Learning for Healthcare Conference 2021
    HTML Code

2019

  1. Model Selection in Bayesian Neural Networks via Horseshoe Priors.
    Ghosh, Soumya,  Yao, Jiayu, and Doshi-Velez, Finale
    JMLR 2019
    HTML

2018

  1. Structured variational learning of Bayesian neural networks with horseshoe priors
    Ghosh, Soumya,  Yao, Jiayu, and Doshi-Velez, Finale
    In International Conference on Machine Learning 2018
    HTML Code

2016

  1. Normal/abnormal heart sound recordings classification using convolutional neural network
    Nilanon, Tanachat,  Yao, Jiayu, Hao, Junheng, Purushotham, Sanjay, and Liu, Yan
    In 2016 computing in cardiology conference (CinC) 2016
    HTML

Workshops and Preprints

2023

  1. Inverse Reinforcement Learning with Multiple Planning Horizons
    Yao, Jiayu, Doshi-Velez, Finale, and Engelhardt, Barbara
    In NeurIPS 2023 Workshop on Generalization in Planning 2023

2022

  1. An Empirical Analysis of the Advantages of Finite-vs Infinite-Width Bayesian Neural Networks
    Yao, Jiayu, Yacoby, Yaniv, Coker, Beau, Pan, Weiwei, and Doshi-Velez, Finale
    arXiv preprint arXiv:2211.09184 2022
    arXiv
  2. A Framework for the Evaluation of Clinical Time Series Models
    Gao, Michael,  Yao, Jiayu, and Henao, Ricardo
    In NeurIPS 2022 Workshop on Learning from Time Series for Health 2022
    HTML
  3. Success of Uncertainty-Aware Deep Models Depends on Data Manifold Geometry
    Penrod, Mark, Termotto, Harrison, Reddy, Varshini,  Yao, Jiayu, Doshi-Velez, Finale, and Pan, Weiwei
    arXiv preprint arXiv:2208.01705 2022
    arXiv
  4. From Soft Trees to Hard Trees: Gains and Losses
    Zeng, Xin,  Yao, Jiayu, Doshi-Velez, Finale, and Pan, Weiwei
    2022
    HTML
  5. Policy Optimization with Sparse Global Contrastive Explanations
    Yao, Jiayu, Parbhoo, Sonali, Pan, Weiwei, and Doshi-Velez, Finale
    arXiv preprint arXiv:2207.06269 2022
    arXiv

2020

  1. Amortised Variational Inference for Hierarchical Mixture Models
    Antorán, Javier,  Yao, Jiayu, Pan, Weiwei, Hernández-Lobato, José Miguel, and Doshi-Velez, Finale
    arXiv preprint arXiv:2111.03144 2020
    arXiv

2019

  1. Output-constrained Bayesian neural networks
    Yang, Wanqian, Lorch, Lars, Graule, Moritz A, Srinivasan, Srivatsan, Suresh, Anirudh,  Yao, Jiayu, Pradier, Melanie F, and Doshi-Velez, Finale
    arXiv preprint arXiv:1905.06287 2019
    arXiv Code
  2. Quality of uncertainty quantification for Bayesian neural network inference
    Yao, Jiayu, Pan, Weiwei, Ghosh, Soumya, and Doshi-Velez, Finale
    arXiv preprint arXiv:1906.09686 2019
    arXiv

2018

  1. Evaluating reinforcement learning algorithms in observational health settings
    Gottesman, Omer, Johansson, Fredrik, Meier, Joshua, Dent, Jack, Lee, Donghun, Srinivasan, Srivatsan, Zhang, Linying, Ding, Yi, Wihl, David, Peng, Xuefeng, and others,
    arXiv preprint arXiv:1805.12298 2018
    arXiv
  2. Direct policy transfer via hidden parameter markov decision processes
    Yao, Jiayu, Killian, Taylor, Konidaris, George, and Doshi-Velez, Finale
    In LLARLA Workshop, FAIM 2018
    HTML
  3. Projected BNNs: Avoiding weight-space pathologies by learning latent representations of neural network weights
    Pradier, Melanie F, Pan, Weiwei,  Yao, Jiayu, Ghosh, Soumya, and Doshi-Velez, Finale
    arXiv preprint arXiv:1811.07006 2018
    arXiv