WAIC

WAIC can refer to more than one concept in different fields. The two most common uses are described below.

Watanabe–Akaike Information Criterion (WAIC) is a Bayesian model comparison criterion used to assess predictive accuracy. It is based on the posterior distribution of the model parameters and provides an estimate of how well a model will predict new data. WAIC uses the log pointwise predictive density and an estimate of the effective number of parameters to adjust for model complexity. Specifically, lppd is the sum of the log predictive densities for each data point under the posterior, and p_waic is the sum of the variances of the log likelihoods across posterior draws. WAIC is computed as -2(lppd - p_waic). Lower WAIC values indicate better expected predictive performance, and the criterion is designed to be asymptotically equivalent to leave-one-out cross-validation under suitable conditions.

World Artificial Intelligence Conference (WAIC) is an international technology conference held in Shanghai, China. It focuses on artificial intelligence research and industry applications, bringing together researchers, policymakers, and executives from academia and industry. The event typically features keynote speeches, panel discussions, exhibitions, and demonstrations of AI-enabled products and services, aiming to showcase advances in AI and foster collaboration across sectors such as finance, healthcare, manufacturing, and smart cities.

In summary, WAIC may denote a statistical model-selection criterion used in Bayesian analysis or a major international AI conference, depending on the context.