High Energy Physics Seminars

Machine learning is now a part of physics for the foreseeable future, but many deep learning tools, architectures, and algorithms are imported from industry to physics with minimal modifications. Does physics really need all of these fancy techniques, or does “dumb” machine learning with the simplest possible neural networks suffice? I will argue that the needs for interpretability and uncertainty quantification in physics applications of machine learning mitigate toward the use of simpler tools with more predictable performance. I will give several examples illustrating how tools imported from physics may be used to better understand the training dynamics of fully-connected networks, and conversely, how the topology and geometry of collider physics data may be used as a testbed for theories of machine learning relevant for data “in the wild”.