Ikibazo gihuza ibibazo · Biragoye Urwego

__AIU_PROTECTED_13 __- Ikibazo cyo Kwiga

Ongera usuzume uburyo meta-yiga itoza imirimo, uburyo MAML ihuza intangiriro yicyitegererezo, nuburyo uburyo bwa prototype butondekanya ingero nkeya.

Inzira zijyanye no kuyoboraMeta Kwiga
Ikibazo 1 Bya 8

Niki kwiga-meta, cyangwa 'kwiga kwiga,' bigamije mbere na mbere kugeraho?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Niki kwiga-meta, cyangwa 'kwiga kwiga,' bigamije mbere na mbere kugeraho?
    • Kumenyera byihuse imirimo mishya uhereye kuburorero buke
    • Gufata mu mutwe dataset imwe nini
    • Kurandura ibikenewe mumahugurwa ayo ari yo yose
    • Gucomeka moderi kugeza kuri zeru
  2. Mugihe cya meta-imyitozo, ni gute amakuru asanzwe atunganijwe?
    • Nkumurimo umwe udafite amacakubiri
    • Nka mirimo myinshi, buriwese hamwe nugushigikira hamwe nibibazo byashizweho
    • Nkuko ishusho imwe yasubiwemo
    • Nkurusaku rutemewe gusa
  3. MAML yiga iki?
    • Ibyiciro bihamye bitigera bihinduka
    • Dataset yibibi
    • Icyitegererezo cyintangarugero gishobora guhuzwa neza nimirimo mishya muntambwe nke
    • Kode yo guhagarika
  4. Nigute imiyoboro ya prototypical itandukanya ingero nshya?
    • Mugukora amashusho mashya
    • Mugufata mumutwe pigiseli indangagaciro
    • Kubitekerezaho
    • Mugereranije na prototypes yishuri yize mumwanya wo gushiramo
  5. Ni uruhe ruhare rw'imbere ya MAML?
    • Guhuza icyitegererezo kumurimo runaka hamwe nintambwe nkeya
    • Guhindura imibare
    • Gusiba ibipimo
    • Gushiraho igipimo cyo kwiga kuri zeru
  6. How does GPT-3 perform the few-shot demonstrations in the cited paper?
    • By updating all model weights after every example
    • By including examples in the prompt without gradient updates or fine-tuning
    • By storing one prototype per class
    • By running a MAML inner loop during inference
  7. In Reptile, how is the meta-initialization updated after adapting on a sampled task?
    • Replace the model with a nearest-neighbor table
    • Average class labels without updating parameters
    • Move the initialization toward the task-adapted parameters
    • Train a separate model for every example
  8. What computational shortcut distinguishes Reptile from the second-order MAML formulation?
    • It requires no gradient steps
    • It stores only class prototypes
    • It trains without tasks or task-specific data
    • It can use a first-order update without computing second-order derivatives