Kedis ku xidhan hagaha · Adag Heerka

Kediska Shabakadda Kaabsoosha

Wuxuu hubiyaa vectors kaabsulka, jiheynta qayb-ilaa-dhan, xog-ururin, buundooyinka heshiiska iyo sababta marin-u-eegiddu u noqon karto mid qaali ah.

Wadooyinka haga ee la xidhiidhaShabakadaha Kaabsoosha
Su'aal 1 ee 8

Shabakadda kaabsoosha, muxuu ka dhigan yahay DHEERARKA vector kaabsoosha kaabsoosha?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Shabakadda kaabsoosha, muxuu ka dhigan yahay DHEERARKA vector kaabsoosha kaabsoosha?
    • Suurtagalnimada in hay'addu ay joogto
    • Midabka muuqaalka la ogaaday
    • Heerka barashada lakabkaas
    • Tirada neurons ee kaabsoosha
  2. Waa maxay dhibaatada haysata CNN-yada caadiga ah ee shabakadaha kaabsal gaar ahaan loogu talagalay in wax laga qabto?
    • Tababar gaabis ah oo ku saabsan GPU-yada
    • Luminta cilaaqaadyada bannaan/qayb-dhan, tusaale; aqbala qaybaha wejiga ee gariirsan
    • Awood la'aanta in lagu farsameeyo sawirada midabka
    • Halbeegyo aad u yar oo la baran karo
  3. Waa maxay magaca algorithm ee go'aaminaya kaabsoosha heerka hoose ah ee ku xidhan kaabsoosha heerka sare ah?
    • Dib u faafinta waqtiga
    • Goynta si tartiib ah
    • Dariiq toosan oo heshiis ah
    • Dufcaddii caadiga ahayd
  4. Yaa soo bandhigay shabakadda kaabsoosha oo leh habayn firfircoon 2017?
    • Yann LeCun iyo asxaabtiisa
    • Ian Goodfellow iyo asxaabtiisa
    • Vaswani iyo asxaabtiisa
    • Sabour, Frosst, iyo Hinton
  5. Waa maxay ujeedada shaqada 'squash' ee shabakada kaabsoosha?
    • Si caadi looga dhigo dhererka vector inta u dhaxaysa 0 iyo 1 iyadoo la ilaalinayo jihadeeda
    • Si loo tirtiro kaabsal aan la isticmaalin
    • Si aad sawirada ugu beddesho miisaan cawl
    • Si loo kordhiyo qoto dheer ee shabakada si toos ah
  6. Which result did the original 2017 CapsNet paper report as evidence for routing-by-agreement?
    • Lower memory use than every CNN on ImageNet
    • Better recognition of highly overlapping digits than a convolutional net
    • No need for a training objective
    • Perfect performance on all affine transformations
  7. In the original capsule representation, what information is carried by the direction of a capsule’s output vector?
    • The probability that the entity is absent
    • The number of routing iterations
    • Instantiation or pose parameters of the represented entity
    • The loss assigned to the previous layer
  8. In routing-by-agreement, what makes a lower capsule’s prediction agree with a higher-level capsule?
    • Their capsule names begin with the same letter
    • Their vector lengths are exactly zero
    • They occur in the same image crop
    • Their predicted and output vectors have a large scalar product