Bootstrapping

التمهيد الذاتي

أسلوب تعلّم يُحدِّث فيه النموذج تقديراته بالاعتماد على تقديرات أخرى لم تُتحقَّق بعد، بدلاً من انتظار النتيجة النهائية الفعلية. رغم أنه يبدو دائرياً، إلا أن كل تقدير لاحق يتضمن معلومات إضافية من خطوة حقيقية واحدة.

A learning method where the model updates its estimates based on other, not-yet-verified estimates rather than waiting for the actual final outcome. Although it seems circular, each successive estimate incorporates one additional step of real-world information.

Also translated asالتعلم الذاتي المرحلي، الاستدلال الدوري من التقديرات، التقدير من التقديرات

First appears in this corpus in: Learning to Predict by the Methods of Temporal Differences (1988)

Appears in these papers

Asynchronous Methods for Deep Reinforcement Learning2016 in the sky ✦
Bagging Predictors1996 in the sky ✦
BLIP-2: Bootstrapping Language-Image Pre-Training with Frozen Image Encoders and Large Language Models2023 in the sky ✦
Decision Transformer: Reinforcement Learning via Sequence Modeling2021 in the sky ✦
Deep Reinforcement Learning with Double Q-Learning2016 in the sky ✦
Gaussian Processes for Machine Learning2006 in the sky ✦
MuZero: Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model2020 in the sky ✦
Prioritized Experience Replay2015 in the sky ✦
Self-Instruct: Aligning Language Models with Self-Generated Instructions2022 in the sky ✦
STaR: Bootstrapping Reasoning With Reasoning2022 in the sky ✦
Temporal Difference Learning and TD-Gammon1995 in the sky ✦
Learning to Predict by the Methods of Temporal Differences1988 in the sky ✦
Learning to Predict by the Methods of Temporal Differences1988 in the sky ✦
Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision2023 in the sky ✦
Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision2023 in the sky ✦