提出一种新梯度估计方法，提升贝叶斯推断采样效率与多样性。
Entropy-regularized Gradient Estimators for Approximate Bayesian Inference
- 基于斯坦因算子度量，优化KL散度与交叉熵的梯度估计
- 在分类任务中实现高质量后验采样，提升模型不确定性建模能力
- 适用于需要不确定性感知的强化学习场景
在数据有限的情况下，有效的不确定性量化对提升现代预测模型的准确性和鲁棒性至关重要。尽管贝叶斯方法在此方面表现优异，但其可扩展性常受挑战。采用近似贝叶斯推断时，如何在计算高效的前提下保证后验样本质量尤为关键。本文通过逼近目标近似分布下Kullback-Leibler（KL）散度与交叉熵的梯度流，并在斯坦因算子诱导的度量空间中进行优化，实现对贝叶斯后验的有效估计，从而生成多样化的样本。在分类任务上进行了实证评估，验证了该方法的有效性，并讨论了其在基于模型的强化学习中使用不确定性感知网络动态模型的应用潜力。
原文摘要 · Abstract (English)
Effective uncertainty quantification is important for training modern predictive models with limited data, enhancing both accuracy and robustness. While Bayesian methods are effective for this purpose, they can be challenging to scale. When employing approximate Bayesian inference, ensuring the quality of samples from the posterior distribution in a computationally efficient manner is essential. This paper addresses the estimation of the Bayesian posterior to generate diverse samples by approximating the gradient flow of the Kullback-Leibler (KL) divergence and the cross entropy of the target approximation under the metric induced by the Stein Operator. It presents empirical evaluations on classification tasks to assess the method's performance and discuss its effectiveness for Model-Based Reinforcement Learning that uses uncertainty-aware network dynamics models.
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。