Counterfactual Prediction for Bundle Treatment

Authors: Hao Zou, Peng Cui, Bo Li, Zheyan Shen, Jianxin Ma, Hongxia Yang, Yue He

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we conduct extensive experiments on both synthetic datasets and real world datasets to demonstrate the advantages of our proposed variational sample re-weighting algorithm.
Researcher Affiliation Collaboration 1Tsinghua University, 2Alibaba Group
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper.
Open Datasets Yes Considering that few datasets contain ground truth of different treatment outcomes, we conduct experiments on both synthetic datasets and datasets from a simulator mimicking recommendation systems in real world to evaluate the effectiveness of our method. and There is a simulation environment about document recommendation 1 in Recsim. 1https://github.com/google-research/recsim/blob/master/recsim/environments/interest_exploration.py
Dataset Splits No The paper mentions creating an 'unbiased testing dataset' by shuffling matches of confounders and treatments but does not specify a distinct validation set or explicit train/validation/test splits.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory, or cloud instance types) used for running its experiments, only mentioning general 'deep neural networks'.
Software Dependencies No The paper does not provide specific ancillary software details, such as library names with version numbers (e.g., Python, PyTorch versions), needed to replicate the experiment.
Experiment Setup Yes In this experiment, we set the confounder dimension d = 10, latent dimension k = 3, the number of one-value bits in treatments s = 5, and the noise variable εy N(0, 0.012). and We fixed the sample size n = 10000, the number of document topics d = 4 and selected documents s = 4.