Kai Germaschewski

John D. Haiducek, Thom R. Edwards, Wade Duvall et al.

The software community has specific definitions for terms such as "open source software," "free software," and "permissive license," but scientists proposing software development efforts to NASA are not always knowledgeable about these definitions. Misunderstandings about the meaning of these terms can result in problems of fairness with solicitations, because scientists who interpret the terms differently than NASA intends may either needlessly limit the scope of their proposed work, or unwittingly propose work that does not comply with software licensing requirements. It is therefore recommended that NASA adopt definitions of the above terms that are in line with software community usage, that these definitions be communicated as part of solicitations to ensure a common understanding, and that proposals be required to identify what software licenses the proposers expect to use.

John Donaghy, Kai Germaschewski

The particle-in-cell numerical method of plasma physics balances a trade-off between computational cost and intrinsic noise. Inference on data produced by these simulations generally consists of binning the data to recover the particle distribution function, from which physical processes may be investigated. In addition to containing noise, the distribution function is temporally dynamic and can be non-gaussian and multi-modal, making the task of modeling it difficult. Here we demonstrate the use of normalizing flows to learn a smooth, tractable approximation to the noisy particle distribution function. We demonstrate that the resulting data driven likelihood conserves relevant physics and may be extended to encapsulate the temporal evolution of the distribution function.