Collaborative Heterogeneous Causal Inference Beyond Meta-analysis

Authors: Tianyu Guo, Sai Praneeth Karimireddy, Michael Jordan

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct the experiment using synthetic dataset. We calculate the mean squared error (MSE) of Meta-IPW, CLB-IPW, and Meta-AIPW, and CLB-AIPW through 2000 Monte Carlo Simulations, with four replications of different {ยต(k)} s. Figure 2 shows the d KL MSE curve.
Researcher Affiliation Academia 1Department of Statistics, UC Berkeley. 2Department of EECS, University of California, UC Berkeley.
Pseudocode Yes Algorithm 1 CLB-IPW Algorithm
Open Source Code No The paper does not explicitly state that source code for the methodology is openly available, nor does it provide a link to a code repository.
Open Datasets Yes Our data comes from two studies about preventing sharing fake news during COVID-19. Roozenbeek et al. (2021) replicates the experiment of Pennycook et al. (2020)...
Dataset Splits No The paper provides sample sizes for synthetic data (N (k) = 1000, 2000, 3000 for source, N (t) = 10000 for target) and refers to real-world studies, but does not specify explicit train/validation/test dataset splits.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running experiments.
Software Dependencies No The paper describes algorithms and methods used (e.g., Fed Avg, SCAFFOLD) but does not list specific software dependencies with version numbers required for replication.
Experiment Setup No The paper describes the models used for estimation (exponential tilting, linear model) and details of synthetic data generation and heterogeneity, but it does not provide specific experimental setup details such as hyperparameters (e.g., learning rates, batch sizes, epochs) for training.