Laajte ci xam-xam kernel tangente neuronal

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

1. Ci yaatuwaayu yaatuwaayu amul àpp, naka la Kernel Tangent Neural di doxalee ci diiru tàggat yaram?
   • Dafay wéy di nekk
   • Jéego bu nekk dafay gaawa soppeeku
   • Dafay nekk nul
   • Dafay màgg bu baax
2. Luy NTK ci wàllu math?
   • Hessian bu ñàkk
   • Produit bi ci biir bu gradient parametru reso bi ngir ñaari duggal
   • softmax bi ci génne yi
   • Variance poid yi
3. Kan moo dugal kernel tangent neuronal bi?
   • Hinton ak Salahutdinov
   • Lecun ak Bengio
   • Jakot, Gabriel, ak Hongler (2018)
   • Vaswani ak ay naataangoom
4. Luy regime 'tàggat tayeel'?
   • Suñu noppee tàggat yaram
   • Sudee optimisatër bi jëfandikoowul momentum
   • Sudee dañuy dugal done yi ndànk
   • Sudee parametre yi deñuy toxu ci ndoorte li, reso bi di doxalee ci anam wu ligneer
5. Benn ci gàttal yi gëna am solo ci theorie NTK ci reso dëgg yi mooy:
   • Jàngale ci reso yu am yaatuwaayu àpp
   • Rolu wàccinu gradient
   • am ay poid
   • Jëfandikoo done tàggat yaram
6. Ndax NTK dafa wàññi tàggat ci kernel regression, lan mooy mëna am?
   • dindi soxla benn done
   • Xalaatal njariñu liñ tàggat te doo tàggat bu baax
   • Modèle yuy daw te amul ay parametre
   • Garanti nul njuumte test saa yu nekk
7. In the infinite-width limit, what probabilistic model are neural networks equivalent to at initialization?
   • Benn xeetu Markov bu nëbbu
   • Ensemble garabu dogal
   • A Gaussian process
   • A finite-state game controller
8. For least-squares regression in the NTK limit, what mathematical form does the network function's training evolution follow?
   • A discrete rule that randomly changes labels
   • A decision tree that grows one branch per epoch
   • A nonlinear search over audio frames
   • A linear differential equation in function space