Papers for

automated experimenters

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Adaptive interventions speed online causal discovery with fixed confidence

FOCUS: Fixed-Confidence Online Causal Learning Using Sequential Adaptive Interventions

Abstract: We study fully online fixed-confidence causal discovery without any historical observational data. Starting from zero samples, the learner sequentially selects interventions to recover both the causal DAG and its edge weights under a linear-Gaussian structural equation model. We establish an instance-dependent lower bound for any $(ε,δ)$-correct algorithm and propose \textsc{FOCUS}, which adaptively allocates interventions through an online max--min game. A key contribution is a computable concentration inequality for the accumulated KL divergence involving causal parameters shared across interventions. We prove that \textsc{FOCUS} is $(ε,δ)$-correct and that its expected stopping time matches the lower bound in its $Θ(\log(1/δ))$ dependence up to an instance-dependent constant. Experiments demonstrate improved structure and edge-weight recovery and confirm the predicted stopping-time trend. Our codes are available on https://anonymous.4open.science/r/FOCUS_code-76E5

Sat 26 SeptMachine Learning

The gist

Figuring out cause-and-effect rules can be tough, especially without data already collected. The authors study how a computer can experiment step-by-step, choosing interventions to quickly and confidently learn these causal relationships from scratch. They provide a method called FOCUS that smartly picks the next experiment based on what it has learned so far. The method is proven to stop after the fewest necessary experiments needed to be confident, matching a theoretical lower bound.

Open → 2609.32165v1