Truly Batch Apprenticeship Learning with Deep Successor Features

Authors: Donghun Lee, Srivatsan Srinivasan, Finale Doshi-Velez

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate our model s superior performance in batch settings with both classical control tasks and a real-world clinical task of sepsis management in the ICU.
Researcher Affiliation Academia Donghun Lee , Srivatsan Srinivasan and Finale Doshi-Velez SEAS, Harvard University {srivatsansrinivasan, donghunlee}@g.harvard.edu, finale@seas.harvard.edu,
Pseudocode Yes Algorithm 1 Batch Max-Margin IRL
Open Source Code Yes Link to our code repository can be found in Section 4 of the Appendix.
Open Datasets Yes The input data was obtained from the Multiparameter Intelligent Monitoring in Intensive Care (MIMIC-III v1.4) database [Johnson et al., 2016] and included a cohort of 17,898 patients fulfilling Sepsis-3 criteria.
Dataset Splits Yes We followed a 70 30 train-validation split in our batch data. AND We then performed a 60-20-20 train-val-test split on our dataset.
Hardware Specification No The paper does not provide specific details on the hardware used for running experiments, such as GPU or CPU models.
Software Dependencies No The paper mentions software like DDQN and OpenAI Gym environments but does not provide specific version numbers for software dependencies or libraries.
Experiment Setup Yes Parametric details of all the models are in Appendix Table 2.