论文标题
量子通勤模型(IA):CHSH游戏和其他示例:最佳状态的独特性
The quantum commuting model (Ia): The CHSH game and other examples: Uniqueness of optimal states
论文作者
论文摘要
储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。
We present in this paper that the CHSH game admits one and only one optimal state and so remove all ambiguity of representations. More precisely, we use the well-known universal description of quantum commuting correlations as state space on the universal algebra for two player games, and so allows us to unambigiously compare quantum strategies as states on this common algebra. As such we find that the CHSH game leaves a single optimal state on this common algebra. In turn passing to any non-minimal Stinespring dilation for this unique optimal state is the only source of ambiguity (including self-testing): More precisely, any state on some operator algebra may be uniquely broken up into its minimal Stinespring dilation as an honest representation for the operator algebra followed by its vector state. Any other Stinespring dilation however arises simply as an extension of the minimal Stinespring dilation (i.e., as an embedding of the minimal Hilbert space into some random ambient one). As such this manifests the only source of ambiguity appearing in most (but not all!) traditional self-testing results such as for the CHSH game as well as in plenty of similar examples. We then further demonstrate the simplicity of our arguments on the Mermin--Peres magic square and magic pentagram game. Most importantly however, we present this article as an illustration of operator algebraic techniques on optimal states and their quotients, and we further pick up the results of the current article in another following one (currently under preparation) to derive a first robust self-testing result in the quantum commuting model.