Falcon

Excited to launch Atinary’s global optimization ML algorithm: Falcon. Which supports: All variables types: Continuous, Discrete & Categorical Pareto optimization (multi-objective) Enable physicochemical descriptors Three alternatives of surrogate models & acquisition functions Atinary™ Falcon is a general-purpose optimization algorithm. It can solve optimization problems that include continuous, discrete and/or categorical variables with or without physicochemical […]

Concept

Excited to launch Atinary’s global optimization ML algorithm: Falcon. Which supports:

  • All variables types: Continuous, Discrete & Categorical
  • Pareto optimization (multi-objective)
  • Enable physicochemical descriptors
  • Three alternatives of surrogate models & acquisition functions

Atinary™ Falcon is a general-purpose optimization algorithm. It can solve optimization problems that include continuous, discrete and/or categorical variables with or without physicochemical descriptors, as well as batch-constrained optimization.

Atinary™ Falcon GPBO uses Gaussian Process Bayesian Optimization as surrogate model. Typically, GPBO is well suited for optimization problems that can potentially be solved with a relatively small number of experiments. However, GPBO scales cubically with the number of experiments. Thus, the computational cost can potentially be very high if used in complex simulation cases.

Atinary™ Falcon DNGO (Deep Network for Global Optimization) maintains desirable properties of the Gaussian Processes (e.g. management of uncertainty) while improving its scalability. Specifically, unlike a standard Gaussian process, DNGO scales linearly with the number of evaluations or experiments. Falcon DNGO creates a robust, scalable, and effective Bayesian optimization system that generalizes across many global optimization problems, for a suitable set of design choices.

By reducing runtime (up to 120x faster than competing algorithms), Falcon not only accelerates materials discovery, but it also minimizes the electricity consumption of data servers.

Share this article:
linkedin
twitter
facebook