Deep neural networks (DNNs) are ubiquitous in computer vision and natural language processing, but suffer from high inference cost.

We compare the model's performance relative to other approaches on diverse regression and Bayesian optimization tasks, including the challenging but common

Although GNNs are powerful tools for these "forward" prediction tasks, few works discuss the "backward" (or inverse) problem defined on trained GNNs.

Existing methods frequently assume causal sufficiency, disregarding the presence of unobserved confounding variables.

KuaiRand Dataset and compare them against KuaiSim to further assess their performance and behavioral differences.

The general recipe for each communication round is composed of three steps.