Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele

Idanwo Ẹkọ Federated

Apply federated model updates, weighted averaging, and secure aggregation, including why keeping raw records local is not a complete privacy guarantee.

Jẹmọ itọsọna awọn ọnaẸkọ Apapo
Ibeere 1 ti 8

Bawo ni ikẹkọ idapọ ṣe ṣe ikẹkọ awoṣe pinpin laisi aarin data olumulo?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. Bawo ni ikẹkọ idapọ ṣe ṣe ikẹkọ awoṣe pinpin laisi aarin data olumulo?
    • Kọ awoṣe kan nipa ikojọpọ data aise ti gbogbo eniyan si olupin aringbungbun kan
    • Lo data sintetiki nikan kii ṣe data gidi rara
    • Kọ awoṣe ti o pin nipa fifiranṣẹ awọn imudojuiwọn awoṣe dipo data aise si olupin kan
    • Ṣiṣe awoṣe lọtọ, ti ko ni ibatan lori ẹrọ kọọkan lailai
  2. Kini Algoridimu Federated (FedAvg) ṣe lori olupin naa?
    • Nparẹ gbogbo awọn awoṣe alabara
    • Ṣe iwọn awọn imudojuiwọn iwuwo lati ọdọ awọn alabara, nigbagbogbo iwuwo nipasẹ iye data
    • Firanṣẹ data aise pada si awọn alabara
    • Ṣe ikẹkọ awoṣe patapata funrararẹ
  3. Ọja wo ni olokiki lo ẹkọ idapọ lati mu awọn asọtẹlẹ dara si?
    • A aarin spam imeeli àlẹmọ
    • Satẹlaiti oju ojo
    • A fidio game console
    • _AIU_PROTECTED_11__'s Gboard keyboard
  4. Why might a hospital collaboration investigate federated training?
    • It can train through local computations without centrally pooling patient records
    • It automatically makes all sharing legally compliant
    • It removes every risk of revealing patient information
    • It requires no evaluation of the resulting model
  5. Kini idi ti awọn imudojuiwọn awoṣe nikan kii ṣe ikọkọ ni kikun laifọwọyi?
    • Awọn imudojuiwọn nigbagbogbo jẹ ariwo laileto
    • Awọn imudojuiwọn tun le jo alaye nipa data abẹlẹ
    • Awọn imudojuiwọn ni awọn aise data ko yipada
    • Awọn imudojuiwọn ko ṣe firanṣẹ lori awọn nẹtiwọọki
  6. Isoro wo ni data 'ti kii ṣe IID' fa ninu ẹkọ ti o ni idapọ?
    • Gbogbo awọn ẹrọ ni data kanna, nitorinaa ko si nkan ti a kọ
    • Awọn olupin nṣiṣẹ jade ti iranti lesekese
    • Awoṣe le kọ ẹkọ awọn aworan nikan
    • Awọn ipinpinpin data awọn alabaṣe yatọ pupọ, ṣiṣe aropin le
  7. Two FedAvg clients return scalar weights 0 and 10 after using 1 and 3 training examples. What is the data-weighted average?
    • 7.5
    • 5
    • 10
    • 2.5
  8. What does a successful secure-aggregation protocol aim to reveal to the federated server?
    • Every client’s raw training record
    • Only a participant’s device password
    • The aggregate update rather than each participant’s individual value
    • A complete guarantee against all information in the aggregate