Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

SyncTwin: Treatment Effect Estimation with Longitudinal Outcomes

Authors: Zhaozhi Qian, Yao Zhang, Ioana Bica, Angela Wood, Mihaela van der Schaar

NeurIPS 2021 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In addition to extensive simulations, we conduct an observational study using real EHR data and successfully reproduced the ๏ฌndings of a randomized controlled clinical trial with Sync Twin.
Researcher Affiliation Collaboration Zhaozhi Qian University of Cambridge EMAIL Yao Zhang University of Cambridge EMAIL Ioana Bica University of Oxford The Alan Turing Institute EMAIL Angela Mary Wood University of Cambridge EMAIL Mihaela van der Schaar University of Cambridge UCLA The Alan Turing Institute EMAIL
Pseudocode Yes The pseudocode is described in A.7.
Open Source Code Yes The implementation of Sync Twin and the experiment code are available at https:// github.com/Zhaozhi QIAN/Sync Twin-Neur IPS-2021 or https://github.com/orgs/ vanderschaarlab/repositories
Open Datasets Yes We used medical records from English National Health Service general practices that contributed anonymised primary care electronic health records to the Clinical Practice Research Datalink (CPRD), covering approximately 6.9 percent of the UK population [25].
Dataset Splits Yes They were split into three equally-sized subsets for training, validation and testing, each with 17,371 treated and 24,557 controls.
Hardware Specification No Our text does not contain specific details about the hardware used to run the experiments, such as exact GPU or CPU models, or memory specifications.
Software Dependencies No Our text does not contain specific version numbers for software dependencies or libraries used in the experiments.
Experiment Setup No Our text does not contain specific experimental setup details such as hyperparameter values (e.g., learning rate, batch size, epochs) or detailed training configurations.