Maswali yanayounganishwa na mwongozo · Ngumu Kiwango

Maswali ya Miundo Kulingana na Nishati

Tumia uwezekano wa msingi wa nishati, utendakazi wa kugawanya, na viwango vya juu vya alama, ikijumuisha jinsi nishati linganifu huamua uwiano wa kawaida wa usambazaji.

Njia za mwongozo zinazohusianaMiundo Kulingana na Nishati
Swali 1 ya 8

Katika modeli inayotegemea nishati, nishati inahusiana vipi na uwezekano wa sehemu ya data?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Katika modeli inayotegemea nishati, nishati inahusiana vipi na uwezekano wa sehemu ya data?
    • Nishati ya juu inamaanisha uwezekano mkubwa
    • Nishati ya chini inamaanisha uwezekano mkubwa (p(x) ni sawia na exp(-E(x)))
    • Nishati na uwezekano havihusiani
    • Uwezekano ni sawa na nishati moja kwa moja
  2. Ni nini hufanya mafunzo ya mifano ya nishati kuwa magumu?
    • Kazi ya nishati haiwezi kuwa mtandao wa neva
    • Zinahitaji data iliyo na lebo pekee
    • Hawawezi kuwakilisha picha
    • Kitendakazi cha kuhesabu Z (kirekebishaji juu ya viingizo vyote) kwa kawaida hakiwezi kukokotwa
  3. Ni njia gani ya sampuli hutumika sana kuchora sampuli kutoka kwa EBM?
    • Mienendo ya Langevin (MCMC inayotegemea gradient)
    • Utafutaji wa boriti
    • Quicksort
    • Pasi moja ya mbele
  4. Kwa nini njia kama kulinganisha alama zinaweza kuzuia kuweka kazi ya kizigeu Z?
    • Kwa sababu Z daima ni sawa na 1
    • Kwa sababu wanapuuza kazi ya nishati
    • Kwa sababu gradient ya logi p(x) ni -gradient ya E(x), na Z (mara kwa mara katika x) hutoka wakati wa kutofautisha.
    • Kwa sababu wanafanya kazi kwenye data pekee
  5. Ni ipi kati ya hizi ni mfano wa msingi wa nishati?
    • Kibadilishaji
    • Mti wa uamuzi
    • Algorithm ya k-njia
    • Mtandao wa 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