OMP 的主要优点是残差与当前解正交。
假设您一次从 $A$(也称为原子)中选择了所有 $k$ 列,并假设 $A$ 是一个过完备基(这或多或少是 OMP 文献中的标准)。k columns from A (also called atoms) at once and let us also presume that A is an overcomplete basis (this is more or less the standard in OMP literature).
现在,使用您的方法,如果与您的测量 $y$ 最相关的原子与 $A$ 中的 $p < k$ 其他原子线性相关,那么您最终将得到一个 $kp$-sparse 信号,因为$p$ 条目或多或少是多余的。当然,同样的论点可以扩展到相关性较低的原子。你也可能很幸运,永远不会看到这种现象。y is linear-dependent with p<k other atoms in A you will end-up with an k−p-sparse signal, because p entries will be more or less redundant. The same argument can of course be extended to less correlated atoms. You might also be lucky and never see the phenomenon.
让我们举同样的例子,但这次使用 OMP。在第一次迭代期间,您将选择与测量值 $y$ 最相关的原子。之后,您计算 $x$ 中的系数,以使新残差与当前测量近似值正交。换句话说,您获得了当前所选原子提供的大部分信息,因此在下一次迭代期间,您很可能会选择一个包含新信息的原子(问问自己在这种情况下线性相关原子会发生什么)。y. After that you compute the coefficient in x such that the new residual is orthogonal to the current measurements approximation. In other words you got the most of the information provided by the currently selected atom so during the next iteration you are very likely to pick an atom that contains fresh information (ask yourself what would happen with linear dependent atoms in this case).
以下是基于 OMP 和 OLS 的原子选择前瞻策略列表,您可能会感兴趣阅读:POMP、LAOLS和POLS。