Causal Modeling for Fairness In Dynamical Systems

Authors: Elliot Creager, David Madras, Toniann Pitassi, Richard Zemel

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

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate empirically that when environment dynamics are unknown, causal reasoning can help utilize observational data to improve off-policy estimation and learning.
Researcher Affiliation Academia 1University of Toronto 2Vector Institute.
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes 1Code at github.com/ecreager/causal-dyna-fair
Open Datasets No We generate observational data from the SCM in Figure 4a, under a MAXPROF threshold policy. We then consider a new policy πτ with per-group thresholds {τj} as its two parameters.
Dataset Splits No We search over the space of two-threshold policies (one threshold per group) to find the policy with the highest off-policy estimate of the objective on a validation set.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU/GPU models, memory, or cloud instance types) used for running the experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library or solver names with version numbers.
Experiment Setup No The paper describes the general experimental setup in terms of the SCMs and policies being evaluated (e.g., MAXPROF vs. EQOPP policies, threshold parameters), but it does not specify concrete hyperparameter values for any learning algorithms (e.g., learning rates, batch sizes), model initialization, or detailed system-level training settings.