Quiz buy jàngale dooleel

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

1. Lan mooy njàngum dooleel bu dëddu ngir delloosi?
   • Fonksioŋu neexal buy leeral doxalinu eksper biñ seetlu
   • Optimisatër bu gëna gaaw ngir wàcci degrade
   • Benn xeetu wane buñ komprime
   • Setu done yuñ etikete
2. Lan moo waral ñuy wax ni jafe-jafe IRL du lu baax wala lu jaxasoo?
   • Ndax wane yi dañuy baax saa yu nekk
   • Ndax anam yu bari yu wuute yu ñuy neexal mën nañu leeral benn doxalin
   • Ndax neexal yi dañu wara nekk lu baaxul saa yu nekk
   • Ndax soxlawul ay done yu xarañ
3. Lan mooy njariñu delloosi neexal ci klonaasu jeffin?
   • Soxlawolen nit ñi
   • Dafay daw saa yu nekk ci diir bu dëgër
   • Dafay jàpp mébet bi ci suuf, di yamale réew yuñu gisul
   • Dafay moytu jëfandikoo benn sàrt
4. naka lay def ba IRL entropi bu gëna mag di saafara jafe-jafey neexal?
   • Di forse neexal yépp ñu yamale
   • Ni ñuy baña bayyi xel ci trajectoire yu suboptimal
   • Soo dindie fonction neexal bi
   • Ni ñuy modele trajectoire bi ni exponentiel ci neexal bi
5. Ginaaw IRL delloosi ab fonction neexal, lan lañuy faral di def ci topp?
   • Njàngalem dooleel buñ miin mingi juddoo ci neexal boobu
   • Neexal bi dañu koy sànni
   • Demonstration yi dañu leen efaase ngir sakkanal palaas
   • Agent bi dafa bàyyi jàng
6. Ci ban senaario la IRL gëna am njariñ?
   • Sudee neexal bi xam nañu ko bu baax
   • Soo bëggee xam ak yamale mébet yi ci ginaaw doxalinu kàngam
   • Sudee amul benn done ci doxalin
   • Soo bëggee jàng jëf yuñ tëral kese
7. Why can a constant zero reward explain many different expert policies?
   • It assigns the highest reward only to expert actions
   • It proves the expert is optimal
   • Every policy receives the same return, so none is preferred by the reward
   • It makes the transition model deterministic
8. In maximum-entropy IRL, how are higher-reward trajectories treated relative to lower-reward ones?
   • They are assigned probability zero
   • They are treated as equally likely regardless of reward
   • They are removed before policy learning
   • They receive higher probability, with probability depending exponentially on total reward