一篇新论文探讨了证据下界（ELBO）与贝叶斯模型选择中的奥卡姆剃刀之间的关系。研究表明，基于ELBO的超参数学习可能导致过拟合，这与偏好更简单模型的奥卡姆剃刀原则相悖。令人惊讶的是，使用证据本身的贝叶斯模型选择有时会偏好过拟合模型，而ELBO则不会。研究结果表明，从业者应警惕降秩假设（在大模型中为便于处理而必需）如何影响模型选择。 AI
影响 强调了大型贝叶斯模型在模型选择中潜在的陷阱，影响了该领域的从业者。
排序理由 关于贝叶斯推断和模型选择的理论方面的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2604.25984v1 Announce Type: new Abstract: The marginal likelihood, also known as the evidence, is regarded as a mathematical embodiment of Occam's razor, enabling model selection that avoids overfitting. The evidence lower bound (ELBO) objective from variational inference h…
The marginal likelihood, also known as the evidence, is regarded as a mathematical embodiment of Occam's razor, enabling model selection that avoids overfitting. The evidence lower bound (ELBO) objective from variational inference has also been used for similar purposes. Prior wo…