Maswali yanayounganishwa na mwongozo · Kati Kiwango

Maswali ya Maelekezo ya Uanachama

Soma makisio ya uanachama wa mafunzo, yenye miundo ya vivuli, urekebishaji wa LiRA, mipaka ya tofauti-faragha, chanya za uwongo, na tathmini ya mashambulizi.

Njia za mwongozo zinazohusianaMashambulizi ya Maelekezo ya Uanachama
Swali 1 ya 8

Ni shambulio gani la makisio ya wanachama linalojaribu kubaini?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Ni shambulio gani la makisio ya wanachama linalojaribu kubaini?
    • Ikiwa rekodi maalum ilikuwa sehemu ya seti ya mafunzo
    • Usahihi wa mfano
    • Usanifu wa mfano
    • Ukubwa wa seti ya mtihani
  2. How can overfitting affect membership-inference exposure in studied models?
    • It proves every training record is identifiable with certainty
    • It removes the need to evaluate attack errors
    • It can increase the behavioral difference between members and comparable nonmembers
    • It makes all outputs independent of training data
  3. Ni mbinu gani ilifanya Shokri ya awali 2017 et al. kutegemea mashambulizi?
    • Ukataji wa gradient
    • Mifano ya kivuli
    • Utafutaji wa boriti
    • Uwekaji alama
  4. Kwa nini makisio ya uanachama yanaweza kuwa na madhara hata bila kufichua yaliyomo kwenye rekodi?
    • Inapunguza mfano
    • Inabadilisha uzito wa mfano
    • Inafuta data ya mafunzo
    • Uanachama wenyewe unaweza kufichua mambo nyeti, kama vile utambuzi
  5. What calibration distinguishes the cited LiRA approach from a single naive loss threshold?
    • A guarantee that every record has the same inherent difficulty
    • Only an increased generation temperature
    • Per-example score distributions estimated from in and out models
    • A requirement that the target reveal its entire training dataset
  6. What does a properly specified differential-privacy guarantee bound?
    • Changes in output probabilities between neighboring datasets
    • A guarantee that no attack can ever make any correct guess
    • A mandatory deletion of all model outputs
    • Only the raw number of files in a training folder
  7. An attack incorrectly flags 10 of 1,000 actual nonmembers as members. What is its false-positive rate on those nonmembers?
    • 10%
    • 0.1%
    • 100%
    • 1%
  8. Why does the LiRA paper emphasize true-positive rate at low false-positive rates?
    • False positives are impossible by definition
    • Average accuracy can conceal whether confidently flagged records are reliable membership evidence
    • Every member and nonmember population must have identical sizes
    • The model’s training accuracy directly gives attack precision