An outlier is a datapoint set apart from a sample population. The outliers theorem in algorithmic information theory states given a computable sampling method, outliers have to appear. We present a simple proof to the outliers theorem, with exponentially improved bounds. We extend the outliers theorem to dynamical systems. Dynamical systems are guaranteed to hit ever larger outlier states with diminishing measure.
翻译:外部线是一个数据点,它与抽样数据组不同。算法信息理论中的外部线理论指出,它给出了一个可计算抽样方法,外部线必须出现。我们向外部线理论组提出一个简单的证明,其界限有指数性改进。我们把外部线理论组扩大到动态系统。动态系统保证用不断缩小的测量来打击更大的外部状态。