arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.03071stat.MLcs.LG
收藏

关于Brenier极分解的神经实现

On a Neural Implementation of Brenier's Polar Factorization

Apple(苹果公司)

Nina Vesseron, Marco Cuturi

AI总结:

本文提出Brenier极分解定理的神经实现,利用输入凸神经网络参数化势函数,通过凸共轭或辅助网络计算保测映射,应用于非凸优化和非对数凹密度采样。

摘要:

1991年,Brenier证明了一个定理,该定理将方阵的极分解(分解为半正定矩阵与酉矩阵的乘积)推广到任意向量场F: R^d→R^d。该定理被称为极分解定理,它指出任何场F都可以通过凸函数u的梯度与保测映射M的复合来恢复,即F=∇u∘M。我们提出了这一深远理论结果的实用实现,并探索其在机器学习中的可能用途。该定理与最优传输(OT)理论密切相关,我们借鉴神经最优传输领域的最新进展,将势函数u参数化为输入凸神经网络。映射M可以通过恒等式M=∇u*∘F使用u的凸共轭u*逐点求值,也可以作为辅助网络进行学习。由于M通常不是单射,我们考虑估计不适定逆映射的额外任务,该逆映射可以使用随机生成器近似原像测度M^{-1}。我们展示了Brenier极分解在非凸优化问题以及非对数凹密度采样中的可能应用。

英文摘要:

In 1991, Brenier proved a theorem that generalizes the polar decomposition for square matrices -- factored as PSD $\times$ unitary -- to any vector field $F:\mathbb{R}^d\rightarrow \mathbb{R}^d$. The theorem, known as the polar factorization theorem, states that any field $F$ can be recovered as the composition of the gradient of a convex function $u$ with a measure-preserving map $M$, namely $F=\nabla u \circ M$. We propose a practical implementation of this far-reaching theoretical result, and explore possible uses within machine learning. The theorem is closely related to optimal transport (OT) theory, and we borrow from recent advances in the field of neural optimal transport to parameterize the potential $u$ as an input convex neural network. The map $M$ can be either evaluated pointwise using $u^*$, the convex conjugate of $u$, through the identity $M=\nabla u^* \circ F$, or learned as an auxiliary network. Because $M$ is, in general, not injective, we consider the additional task of estimating the ill-posed inverse map that can approximate the pre-image measure $M^{-1}$ using a stochastic generator. We illustrate possible applications of Brenier's polar factorization to non-convex optimization problems, as well as sampling of densities that are not log-concave.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑