Fast, robust approximate message passing
This provides a robustification method for AMP algorithms in optimization problems, which is incremental but addresses a specific vulnerability in existing methods.
The paper tackles the problem of making approximate message passing (AMP) algorithms robust to small corruptions in input data, by introducing a fast spectral pre-processing step that ensures the output remains close to the uncorrupted solution, with error bounds that vanish as corruption size decreases.
We give a fast, spectral procedure for implementing approximate-message passing (AMP) algorithms robustly. For any quadratic optimization problem over symmetric matrices $X$ with independent subgaussian entries, and any separable AMP algorithm $\mathcal A$, our algorithm performs a spectral pre-processing step and then mildly modifies the iterates of $\mathcal A$. If given the perturbed input $X + E \in \mathbb R^{n \times n}$ for any $E$ supported on a $\varepsilon n \times \varepsilon n$ principal minor, our algorithm outputs a solution $\hat v$ which is guaranteed to be close to the output of $\mathcal A$ on the uncorrupted $X$, with $\|\mathcal A(X) - \hat v\|_2 \le f(\varepsilon) \|\mathcal A(X)\|_2$ where $f(\varepsilon) \to 0$ as $\varepsilon \to 0$ depending only on $\varepsilon$.