The source code for the paper ‘Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learning,’ which has been accepted for publication at NeurIPS 2024, is available in this repository
python sim_toy.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0
python sim_tumor.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0
python MCTrue_toy.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0 --MC 10000
python MCTrue_tumor.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0 --MC 10000
python tune_toy.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0 --method TWD
python tune_tumor.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0 --method TWD
python toy_eval.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0 --method TWD
python tumor_eval.py --d_seed 11 --d_number 1000 --e_degree 1.0 --c_degree 1.0 --method TWD
I will continue to update the code over the next few days. please contract 24121534R@connect.polyu.hk if you have any questions