Abstract:
|
Delta smelt are a small endangered fish whose fate is intimately linked with water management practice in the Sacramento river delta system, and who more broadly serve as a barometer for environmental health in the San Francisco Bay. Researchers have developed a stochastic, agent-based simulator to virtualize the system, with the goal of assisting in a study of delta smelt life cycles and to understand sensitivities to myriad natural variables and human interventions. However, the input configuration space is high-dimensional, running the simulator is time-consuming, and its noisy outputs change nonlinearly in both mean and variance. Getting enough runs to effectively learn input-output dynamics requires both a nimble modeling strategy and a parallel supercomputer evaluation. We propose a batch sequential design scheme, generalizing one-at-a-time variance-based active learning for hetGP surrogates, as a means of keeping multi-core cluster nodes fully engaged with expensive runs. Our acquisition strategy is carefully engineered to favor selection of replicates which boost statistical and computational efficiencies when training surrogates to isolate signal in high noise regions.
|