Idanwo Awọn awoṣe orisun Agbara

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

1. Ninu awoṣe ti o da lori agbara, bawo ni agbara ṣe ni ibatan si iṣeeṣe aaye data kan?
   • Agbara giga tumọ si iṣeeṣe giga
   • Agbara kekere tumọ si iṣeeṣe giga (p(x) ni ibamu si exp(-E(x))))
   • Agbara ati iṣeeṣe ko ni ibatan
   • Iṣeeṣe dọgba si agbara taara
2. Kini o jẹ ki awọn awoṣe ti o da lori agbara ikẹkọ nira?
   • Iṣẹ agbara ko le jẹ nẹtiwọọki nkankikan
   • Wọn nilo data ike nikan
   • Wọn ko le ṣe aṣoju awọn aworan
   • Iṣẹ ipin Z (normalizer lori gbogbo awọn igbewọle) jẹ aiṣedeede nigbagbogbo lati ṣe iṣiro
3. Ọna iṣapẹẹrẹ wo ni a lo nigbagbogbo lati fa awọn ayẹwo lati EBM kan?
   • Awọn ìmúdàgba Langevin (MCMC ti o da lori gradient)
   • Iwadi ina
   • Quicksọtọ
   • A nikan siwaju kọja
4. Kini idi ti awọn ọna bii ibaramu Dimegilio yago fun ṣiṣe iṣiro iṣẹ ipin Z?
   • Nitori Z nigbagbogbo dogba si 1
   • Nitoripe wọn foju iṣẹ agbara
   • Nitoripe gradient ti log p(x) jẹ -gradient ti E(x), ati Z (iduroṣinṣin ni x) ṣubu jade nigbati o ba ṣe iyatọ
   • Nitoripe wọn ṣiṣẹ nikan lori data ọtọtọ
5. Ewo ninu iwọnyi jẹ awoṣe ti o da lori agbara?
   • Amunawa
   • Igi ipinnu
   • Awọn k-tumosi alugoridimu
   • Nẹtiwọọki Hopfield
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