Random forest-based prediction of decay modes and half-lives of superheavy nuclei

In 2011, 100 new nuclides were discovered. They joined the approximately 3,000 stable and radioactive nuclides that either occur naturally on Earth or are synthesized in the laboratory. Every atomic nucleus, characterized by a specific number of protons and neutrons, occupies a spot on the chart of nuclides, which is bounded by 'drip lines' indicating the values of neutron and proton number at which nuclear binding ends. The placement of the neutron drip line for the heavier elements is based on theoretical predictions using extreme extrapolations, and so is uncertain. However, it is not known how uncertain it is or how many protons and neutrons can be bound in a nucleus. Here we estimate these limits of the nuclear 'landscape' and provide statistical and systematic uncertainties for our predictions. We use nuclear density functional theory, several Skyrme interactions and high-performance computing, and find that the number of bound nuclides with between 2 and 120 protons is around 7,000. We find that extrapolations for drip-line positions and selected nuclear properties, including neutron separation energies relevant to astrophysical processes, are very consistent between the models used.

Machine Learning (ML) is one of the most exciting and dynamic areas of modern research and application. The purpose of this review is to provide an introduction to the core concepts and tools of machine learning in a manner easily understood and intuitive to physicists. The review begins by covering fundamental concepts in ML and modern statistics such as the bias-variance tradeoff, overfitting, regularization, generalization, and gradient descent before moving on to more advanced topics in both supervised and unsupervised learning. Topics covered in the review include ensemble models, deep learning and neural networks, clustering and data visualization, energy-based models (including MaxEnt models and Restricted Boltzmann Machines), and variational methods. Throughout, we emphasize the many natural connections between ML and statistical physics. A notable aspect of the review is the use of Python Jupyter notebooks to introduce modern ML/statistical packages to readers using physics-inspired datasets (the Ising Model and Monte-Carlo simulations of supersymmetric decays of proton-proton collisions). We conclude with an extended outlook discussing possible uses of machine learning for furthering our understanding of the physical world as well as open problems in ML where physicists may be able to contribute.