Kedis ku xidhan hagaha · Dhexdhexaad Heerka

Hababka Jilaa-Critic Quiz

Waxay hubisaa jilayaasha iyo doorarka dhaleeceynta, faa'iidada iyo calaamadaha TD, yaraynta kala duwanaanshaha, falalka joogtada ah iyo tusaalooyinka PPO. dhaqanka casriga ah.

Wadooyinka haga ee la xidhiidhaHababka Dhaliilaha Jilaaga
Su'aal 1 ee 8

Habka Jilaaga-Critic, waa maxay doorka 'dhaleeceeyaha'?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Habka Jilaaga-Critic, waa maxay doorka 'dhaleeceeyaha'?
    • Si loo qiyaaso qiimaha gobolada (ama ficilada) oo la qiimeeyo doorashada jilaaga
    • Si loo doorto tallaabada la qaadayo
    • Si loo kaydiyo kaydka ku celiska ah
    • In si aan kala sooc lahayn loo baadho deegaanka
  2. Maxay 'faa'iidada' A(s,a) = Q(s,a) - V(-yada) cabbiraysaa?
    • Abaalmarinta qaybta guud
    • Intee in le'eg ayay ficilku ka fiican yahay celceliska qiimaha gobolka
    • Heerka waxbarashada
    • Tirada ficilada la heli karo
  3. Maxay ku darista dhaleecayn u dhimaysaa kala duwanaanshiyaha marka la barbar dhigo hababka saafiga ah ee siyaasad-sii-xoojinta sida REINFORCE?
    • Waxay ka saartaa abaal-marinnada isla'egta
    • Waxay ka dhigaysaa siyaasadda mid go'aaminaysa
    • Dhaliiluhu waxa uu bixiyaa gundhig, markaa cusboonaysiinta waxay isticmaalaan calaamad faa'iido yar oo buuq ah
    • Waxay kordhisaa qodobka dhimista
  4. Awooddee ayay hababka Jilaaga-Critic leeyihiin hababkaas qiimaha leh ee ku salaysan sida Q-Learning?
    • Ficilada kala duwan
    • Isticmaalka abaal-marinnada
    • Dhimista mustaqbalka
    • Goobaha waxqabad ee joogtada ah
  5. Midkee ka mid ah kuwan soo socdaa algorithm-ku-saleysan Jilaa-Critic oo si weyn loogu isticmaalo RLHF ee moodooyinka luqadda?
    • PPO (U-habaynta Siyaasadda Ku-dhawaad)
    • K-macneheedu waa
    • Dib u noqoshada toosan
    • Q-Barashada
  6. Calaamadee uu jilaagu caadi ahaan u isticmaalo si uu u cusboonaysiiyo siyaasadiisa?
    • Gelida pixel ceeriin
    • Faa'iidada ama qaladka TD ee uu bixiyay naqdigu
    • Cabbirka dib u daawaynta
    • Nambar aan sugnayn oo go'an
  7. An action has a positive advantage estimate. How does a policy-gradient actor typically use that signal?
    • Decrease its probability in every state
    • Set the critic’s value estimate to zero
    • Increase the probability of choosing that action in the state
    • Replace the policy with a Q-table
  8. An action’s estimated advantage is negative. What direction should the actor update favor?
    • Increase it because the critic was surprised
    • Delete the state from the environment
    • Raise the reward discount factor automatically
    • Reduce the likelihood of that action relative to the policy baseline