What is Gradient Descent? Definition | AI Understanding

Maanaam

Pexem optimisation buy yeesal paramet yi ci wàll wi wàññi njuumte.

Term yu ci méngoo

DPO (Optimisation préférence directe)

Xeetu tàggat yaram buy gëna suqali xeetu model yi ci ñaari tànneef te soxlawul xeetu neexal bu wuute.

Vecteur buy wane ba ñaata la parametre bu nekk wara soppi ngir wàññi perte.

Tasaaro ci ginaaw

Algorithm bi gëna am solo biy yeesal diisaayu model bi ci tasaare njuumti yiñ wax luy waaja am ci reso bi.

Modèlu njuumte bu dëppoo wala ñàkka yamale ci done wala ci doxalinu model.

Pexem regularisasioŋ buy yombal etiket yu dëgër yi ngir gëna mëna yamale.

Taux de apprentissage

Hyperparametre tàggat yaram buy saytu ñaata parametre di soppi jéego bu nekk ci yeesal.

Nekk leen ci sunuy tegtal yu amul fayda

Gradient Accumulation Stochastic Gradient Descent with Momentum Nesterov Accelerated Gradient Vanishing and Exploding Gradients

Xoolal itam

Ground Truth Generative AI Guardrails Generalization Hallucination GAN (Generative Adversarial Network)Human-in-the-Loop Function Calling

Xoolal Glossaire IA bi yépp Saytu bépp tegtal IA bu amul fayda