The Department of Energy’s SLAC National Accelerator Laboratory has been able to improve the performance of their accelerators by training machine learning algorithms on physics models. Normally these algorithms would need to be trained on vast amounts of pre-existing data; however by teaching these algorithms enough about the physics behind an accelerator, researchers were able to dramatically decrease the amount of pre-existing data needed. Read more about this in today’s #friday-factoid! AI learns physics to optimize particle accelerator performance | SLAC National Accelerator Laboratory