Machine Learning for Health symposium 2023 -- Findings track

no code implementations • 1 Dec 2023 • Stefan Hegselmann, Antonio Parziale, Divya Shanmugam, Shengpu Tang, Mercy Nyamewaa Asiedu, Serina Chang, Thomas Hartvigsen, Harvineet Singh

A collection of the accepted Findings papers that were presented at the 3rd Machine Learning for Health symposium (ML4H 2023), which was held on December 10, 2023, in New Orleans, Louisiana, USA.

Counterfactual-Augmented Importance Sampling for Semi-Offline Policy Evaluation

1 code implementation • NeurIPS 2023 • Shengpu Tang, Jenna Wiens

In applying reinforcement learning (RL) to high-stakes domains, quantitative and qualitative evaluation using observational data can help practitioners understand the generalization performance of new policies.

counterfactual Off-policy evaluation

Leveraging Factored Action Spaces for Off-Policy Evaluation

1 code implementation • 13 Jul 2023 • Aaman Rebello, Shengpu Tang, Jenna Wiens, Sonali Parbhoo

In this work, we propose a new family of "decomposed" importance sampling (IS) estimators based on factored action spaces.

counterfactual Off-policy evaluation

Leveraging Factored Action Spaces for Efficient Offline Reinforcement Learning in Healthcare

2 code implementations • 2 May 2023 • Shengpu Tang, Maggie Makar, Michael W. Sjoding, Finale Doshi-Velez, Jenna Wiens

We study the theoretical properties of our approach, identifying scenarios where it is guaranteed to lead to zero bias when used to approximate the Q-function.

Offline RL reinforcement-learning +1

Machine Learning for Health symposium 2022 -- Extended Abstract track

no code implementations • 28 Nov 2022 • Antonio Parziale, Monica Agrawal, Shalmali Joshi, Irene Y. Chen, Shengpu Tang, Luis Oala, Adarsh Subbaswamy

A collection of the extended abstracts that were presented at the 2nd Machine Learning for Health symposium (ML4H 2022), which was held both virtually and in person on November 28, 2022, in New Orleans, Louisiana, USA.

Towards Data-Driven Offline Simulations for Online Reinforcement Learning

1 code implementation • 14 Nov 2022 • Shengpu Tang, Felipe Vieira Frujeri, Dipendra Misra, Alex Lamb, John Langford, Paul Mineiro, Sebastian Kochman

Modern decision-making systems, from robots to web recommendation engines, are expected to adapt: to user preferences, changing circumstances or even new tasks.

Decision Making reinforcement-learning +1

Model Selection for Offline Reinforcement Learning: Practical Considerations for Healthcare Settings

1 code implementation • 23 Jul 2021 • Shengpu Tang, Jenna Wiens

In this work, we investigate a model selection pipeline for offline RL that relies on off-policy evaluation (OPE) as a proxy for validation performance.

Computational Efficiency Decision Making +4

Clinician-in-the-Loop Decision Making: Reinforcement Learning with Near-Optimal Set-Valued Policies

1 code implementation • ICML 2020 • Shengpu Tang, Aditya Modi, Michael W. Sjoding, Jenna Wiens

We analyze the theoretical properties of the proposed algorithm, providing optimality guarantees and demonstrate our approach on simulated environments and a real clinical task.

Decision Making Reinforcement Learning (RL)

Relaxed Parameter Sharing: Effectively Modeling Time-Varying Relationships in Clinical Time-Series

1 code implementation • Machine Learning for Healthcare 2019 2019 • Jeeheh Oh, Jiaxuan Wang, Shengpu Tang, Michael Sjoding, Jenna Wiens

In settings with limited data, relaxed parameter sharing can lead to improved patient risk stratification performance.

Respiratory Failure Time Series +1