Bayesian Yakadzika Yekudzidza Quiz

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

1. Ko Bayesian neural network inobata sei uremu hwayo zvakasiyana kubva kune yakajairwa?
   • Inobata huremu hwega hwega sechipo chekugovera pane imwe chete yakatarwa kukosha
   • Inoshandisa huremu hwakakwana pachinzvimbo chegumi
   • Inobvisa uremu zvachose
   • Inogovanisa huremu humwe pazvikamu zvese
2. What can a Bayesian approach provide that a single fixed-weight network does not directly represent?
   • A guarantee that every confidence score is calibrated
   • A distribution of predictions that reflects uncertainty about model weights
   • A proof that the model will detect every unfamiliar input
   • A way to remove all noise from the data
3. How can epistemic uncertainty change as relevant, representative training data are added?
   • It is guaranteed to reach zero with any added data
   • It is identical to label noise
   • It may decrease under the chosen model and data assumptions
   • It always increases linearly
4. Sei fungidziro senge musiyano wekufungidzira kana MCMC ichidikanwa?
   • Nekuti neural network haina kurasikirwa nebasa
   • Nekuti maGPU haakwanise kuwedzera
   • Nokuti huremu hunofanira kuva nhamba yakakwana
   • Nekuti iyo chaiyo kumashure kwemamiriyoni ehuremu haigoneke
5. Chii chinoitwa neMonte Carlo kusiya chikoro kufungidzira kusagadzikana?
   • Inochengeta inodonha ichishanda panguva yekuyedza uye inomhanyisa network kakawanda, ichishandisa kupararira kwezvinobuda
   • Inodzima kusiya chikoro uye kudzidzisa kwenguva refu
   • Inobvisa ma neurons zvachose
   • Inotsiva kurasikirwa neruzha rwakangoitika
6. Variational inference inogadziridza posterior estimation serudzii rwedambudziko?
   • Dambudziko rekuronga
   • Dambudziko rekugadzirisa rinoderedza kusiyana kweKL (kuburikidza neELBO)
   • A pure random sampling dambudziko pasina chinangwa
   • Dambudziko rekugadzira hardware
7. How does Bayes’ rule relate a prior, likelihood, and posterior?
   • The posterior is the prior plus the likelihood, without normalization
   • It is the likelihood divided by the prior without normalization
   • The posterior is proportional to the prior times the likelihood
   • It is the fixed point estimate of the most likely weight vector
8. How does MCMC approximate a Bayesian neural-network posterior?
   • By training a single point estimate with dropout disabled
   • By ranking model outputs with a fixed rule
   • By minimizing an ELBO with a simpler variational family
   • By drawing samples through a Markov chain whose stationary distribution targets the posterior