Reparameterization Trick

حيلة إعادة البَرمَتَة

تقنية تنقل العشوائية إلى خارج المخطط الحسابي بتمثيل العينة z كدالة حتمية z = μ + σ·ε حيث ε ضوضاء خارجية، مما يُمكِّن التحديث التراجعي عبر خطوة السحب العشوائي.

A technique that moves randomness outside the computation graph by expressing the sample z as a deterministic function z = μ + σ·ε where ε is external noise, enabling backpropagation through the sampling step.

Also translated asخدعة إعادة الصياغة المعاملية، حيلة إعادة المعلمة، إعادة البرمتة

First appears in this corpus in: Auto-Encoding Variational Bayes (2013)

Appears in these papers

beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework2017 in the sky ✦
Denoising Diffusion Probabilistic Models2020 in the sky ✦
Denoising Diffusion Probabilistic Models2020 in the sky ✦
Variational Inference with Normalizing Flows2015 in the sky ✦
Variational Inference with Normalizing Flows2015 in the sky ✦
Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor2018 in the sky ✦
Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor2018 in the sky ✦
Auto-Encoding Variational Bayes2013 in the sky ✦
Auto-Encoding Variational Bayes2013 in the sky ✦