Fig. 1: Algorithmic framework for a high-throughput experimental loop.
From: Two-step machine learning enables optimized nanoparticle synthesis
The two-step optimization algorithmic framework (blue box) consists of a first HTE loop (runs 2–5) in which the BO (blue points) is sampling the parameter space to train a DNN and a second loop (runs 6–8) in which the DNN (orange points) is allowed to sample the parameter space to validate its regression function. The conditions suggested by the BO and the DNN are tested on a droplet-based microfluidic platform. The absorbance spectrum of each droplet is measured and compared to the target spectrum through the loss function before feeding the BO, while the fully resolved absorbance spectrum is provided to the DNN.
