Kediska-barashada

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

1. Maxay u taagan tahay Q-qiimaha Q(s, a)?
   • Abaalmarinta muddada dheer ee la filayo ee ah qaadista tallaabo ee gobolka, ka dibna u dhaqmo si wanaagsan
   • Suurtagalnimada joogitaanka gobolka s
   • Inta jeer ee falka a la tijaabiyay
   • Abaalmarinta degdega ah ee ficilka a
2. Waa maxay sababta Q-Learning loogu tilmaamay 'siyaasad ka baxsan'?
   • Waxay u baahan tahay qaab deegaan
   • Waxay baran kartaa siyaasadda ugu wanaagsan iyadoo raacaysa siyaasad ka duwan, siyaasad sahaminta
   • Waxa ay ku shaqaysaa qad la'aanta xogta kaydsan
   • Waxay iska indhatirtay gabi ahaanba abaal-marinaha
3. Xeerka cusboonaysiinta, waa maxay sababta dhimista gamma ay maamusho?
   • Heerka sahaminta
   • Cabbirka miiska gobolka
   • Intee in le'eg abaal-marin mustaqbalka ayaa la miisaamaa marka loo eego kuwa dhow
   • Tirada qaybaha
4. Waa maxay farqiga ku meel gaarka ah (TD) ee ku jira Q-Learning?
   • Farqiga u dhexeeya laba gobol
   • Abaalmarinta guud ee dhacdo
   • Waqtiyada heerka waxbarashada gamma
   • r + gamma*max Q(s',a') laga jaray qiyaasta Q(s,a) hadda
5. Waa maxay sababta ay Q-Learning-ka cad ula halgamayso dhibaatooyinka waaweyn sida ciyaaraha fiidyaha ee pixels?
   • Tirada gobolada kala duwan ayaa aad u weyn in lagu kaydiyo miis
   • Ma isticmaali karto abaal-marin
   • Kaliya waxay ku shaqeysaa laba fal
   • Gamma waa inuu ahaadaa 1
6. Waa maxay ujeedada doorashada ficil-damaceed ee epsilon?
   • Si aad had iyo jeer u doorato tallaabada ugu saraysa ee Q-qiimaha
   • Si loo miisaamo sahaminta falalka cusub iyo ka faa'iidaysiga kuwa wanaagsan ee la yaqaan
   • Si loo dejiyo qodobka dhimista
   • Si loo kaydiyo miiska Q
7. What exploration condition appears in the tabular Q-learning convergence result?
   • Only the greedy action is ever sampled
   • Each state must be visited exactly once
   • Every state-action pair must be visited repeatedly under suitable learning conditions
   • The transition model must be known in advance
8. Why is Q-learning called model-free?
   • It never observes the environment
   • It does not use rewards
   • It requires a complete transition table before learning
   • It updates values from sampled rewards and next states without storing transition probabilities