指出用MAML训练神经网络无法真正注入贝叶斯先验，效果不如真实贝叶斯模型。
Comment on "Modeling rapid language learning by distilling Bayesian priors into artificial neural networks"
- MAML仅初始化权重，未在目标函数中真正引入先验
- 模型过拟合严重，对新数据泛化能力差
- 适合关注贝叶斯学习与神经网络融合的学者
McCoy & Griffiths（2025，简称M&G）提出可通过模型无关元学习（MAML）将贝叶斯先验融入人工神经网络（ANN），并以元训练网络在形式语言学习上表现接近Yang & Piantadosi（2023）的贝叶斯学习者为证据，显著优于标准ANN。本文指出，在标准先验定义下，M&G的方法并未真正注入先验，仅实现有益权重初始化，目标函数未改变。若采用更宽松的解释，认为系统整体可视为贝叶斯学习者，仍面临非平凡挑战。进一步分析表明，与真实贝叶斯学习者相比，M&G模型存在严重过拟合，对未见数据泛化性能不佳。
原文摘要 · Abstract (English)
McCoy & Griffiths (2025, henceforth M&G) suggest that a Bayesian prior can be distilled into Artificial Neural Networks (ANNs) through Model-Agnostic Meta-Learning (MAML, Finn et al., 2017). They support this empirically by showing that meta-trained networks demonstrate formal language learning abilities comparable to Yang & Piantadosi (2023)'s Bayesian learner, significantly outperforming standard ANNs. We point out that under the standard interpretation of a prior, M&G's procedure does not actually instill one; it merely initializes network weights favorably, leaving the objective function unchanged. We then consider a more permissive interpretation, where the system as a whole can be seen as implementing a Bayesian learner even without an explicit prior in the objective. We show that this interpretation faces nontrivial challenges. Finally, we assess how well MAML approximates the empirical results of Bayesian learning, showing that unlike genuine Bayesian learners, M&G's model overfits and generalizes poorly to unseen data.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。