Guide-yakabatana mibvunzo · Zvakaoma Nhanho

Energy-Based Models Quiz

Shandisa simba-based probabilities, partition function, uye zvibodzwa gradients, kusanganisira kuti simba rehukama rinosarudza sei zvakajairwa kugovera reshiyo.

Nzira dzekutungamira dzinoenderanaEnergy Based Models
Mubvunzo 1 ye 8

Mune imwe simba-yakavakirwa modhi, simba rinoenderana sei nemukana weiyo data point?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Mune imwe simba-yakavakirwa modhi, simba rinoenderana sei nemukana weiyo data point?
    • Simba guru rinoreva mukana mukuru
    • Simba rakaderera rinoreva mukana mukuru (p(x) inoenzanirana ne exp(-E(x)))
    • Simba uye zvingangoitika hazvina hukama
    • Zvichida zvakaenzana nesimba racho zvakananga
  2. Chii chinoita kuti kudzidzisa mamodheru-akavakirwa simba kunetse?
    • Basa resimba harigone kuve neural network
    • Vanoda data rakanyorwa chete
    • Havagoni kumiririra mifananidzo
    • Iyo yekuparadzanisa basa Z (normalizer pamusoro pezvese zvinopinda) kazhinji haigone kuverengera
  3. Ndeipi nzira yesampling inowanzoshandiswa kudhirowa masampuli kubva kuEBM?
    • Langevin dynamics (gradient-based MCMC)
    • Beam search
    • Quicksort
    • Kupfuura kamwe chete
  4. Sei nzira dzakaita senge chibodzwa kuenzanisa kudzivirira komputa iyo yekuparadzanisa basa Z?
    • Nekuti Z inogara yakaenzana na1
    • Nokuti vanofuratira basa resimba
    • Nekuti kupenya kwegiyo p(x) ndiko -gradient yeE(x), uye Z (inogara mu x) inodonha kana ichisiyanisa.
    • Nekuti ivo vanongoshanda pane discrete data
  5. Ndeipi yeiyi ndiyo yekare simba-yakavakirwa modhi?
    • The Transformer
    • Muti wesarudzo
    • Iyo k-inoreva algorithm
    • Iyo Hopfield network
  6. For a differentiable EBM density with Z constant in x, how does its score relate to energy?
    • The score equals the partition function itself
    • The score equals the negative input gradient of energy
    • The score ignores the learned energy entirely
    • The score is always the positive energy gradient
  7. An EBM adds the same constant c to every E(x). What happens to normalized probabilities?
    • Every probability increases by c
    • All inputs become equally probable
    • They remain unchanged after the partition function is adjusted
    • The normalizer must be set to zero
  8. For two configurations with E(a) lower than E(b), what follows under the same normalized Gibbs model?
    • Configuration a has greater probability than b
    • Configuration b has greater probability because its energy is higher
    • Both probabilities must be exactly one
    • The comparison requires different partition functions for each point