New projected LMC model offers efficient multitask Gaussian process computation

A new paper introduces the "projected LMC" model, an efficient and exact computation method for the Linear Model of Co-regionalization (LMC). This model, a general multitask Gaussian process, typically suffers from cubic complexity. The projected LMC significantly reduces this complexity to be linear in the number of latent processes, provided a mild hypothesis on the noise model is met. This advancement makes the LMC a more competitive and interpretable alternative to existing multitask Gaussian process models, potentially facilitating its adoption in industries like multitask Bayesian optimization. AI

IMPACT This research could lead to more efficient and interpretable multitask Bayesian optimization, potentially accelerating adoption in various industries.

RANK_REASON The cluster contains an academic paper detailing a new computational method for a statistical model. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv stat.ML →

• Gaussian process
• Linear Model of Co-regionalization
• Olivier Truffinet
• projected LMC

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →