The Modern Mathematics of Deep Learningexplained for non-mathematicians

Natural sciences

Scientific applications use known structure as well as observed data.

Lessons

  1. Regularized inverse problem
  2. Eigenvalue equation
  3. Inverse problem objective
  4. Hamiltonian potential terms
  5. Eigenvalue equation
  6. Swapping particles
  7. Finite Rayleigh quotient and local energy
  8. Spin-block determinants
  9. Permutation-equivariant features
  10. Distance-based embeddings
  11. Determinants with learned factors
  12. Embedding into a feature network
  13. A product of orbitals
  14. A finite inverse minimum
  15. Learned inverse reconstruction
  16. Particle and nucleus coordinates
  17. Three learned inverse components
  18. Wavefunction input and output
  19. Separate spin-group embeddings
  20. An undefined source symbol