Automated Hyperparameter Selection for the PC Algorithm
This work addresses a bottleneck in causal inference for researchers and practitioners by automating hyperparameter selection in an unsupervised setting, though it is incremental as it builds on the existing PC algorithm.
The authors tackled the problem of selecting the hyperparameter α for the PC algorithm, which is unsupervised and cannot use traditional cross-validation, by proposing AutoPC, a fast procedure that optimizes α for a user-chosen metric and uses stability between two runs to select the final output, resulting in consistent outperformance of the state of the art across multiple metrics.
The PC algorithm infers causal relations using conditional independence tests that require a pre-specified Type I $α$ level. PC is however unsupervised, so we cannot tune $α$ using traditional cross-validation. We therefore propose AutoPC, a fast procedure that optimizes $α$ directly for a user chosen metric. We in particular force PC to double check its output by executing a second run on the recovered graph. We choose the final output as the one which maximizes stability between the two runs. AutoPC consistently outperforms the state of the art across multiple metrics.