HIGHLIGHTS
- who: Krish Desai and colleagues from the DepartmentUniversity of California, Berkeley, California, USA have published the article: Symmetry discovery with deep learning, in the Journal: (JOURNAL) of 26/05/2022
- what: The authors provide a rigorous statistical definition of the symmetries of a dataset which involves inertial reference densities in analogy to inertial frames in classical mechanics. The deep learning setup is simpler than existing approaches and the authors develop an analytic understanding of the algorithm`s performance in simple cases. While symmetries acting on individual data elements are interesting, they are not the focus . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.