Local Identification of Overcomplete Dictionaries
It provides theoretical guarantees for dictionary learning in signal processing, addressing a fundamental challenge with overcomplete representations, though it is incremental in extending known methods.
This paper tackles the problem of identifying overcomplete dictionaries from noisy training signals, showing that stable local recovery is possible with sparsity levels up to O(μ^{-2}) and signal-to-noise ratios up to O(√d), using a new maximization criterion and an efficient algorithm.
This paper presents the first theoretical results showing that stable identification of overcomplete $μ$-coherent dictionaries $Φ\in \mathbb{R}^{d\times K}$ is locally possible from training signals with sparsity levels $S$ up to the order $O(μ^{-2})$ and signal to noise ratios up to $O(\sqrt{d})$. In particular the dictionary is recoverable as the local maximum of a new maximisation criterion that generalises the K-means criterion. For this maximisation criterion results for asymptotic exact recovery for sparsity levels up to $O(μ^{-1})$ and stable recovery for sparsity levels up to $O(μ^{-2})$ as well as signal to noise ratios up to $O(\sqrt{d})$ are provided. These asymptotic results translate to finite sample size recovery results with high probability as long as the sample size $N$ scales as $O(K^3dS \tilde \varepsilon^{-2})$, where the recovery precision $\tilde \varepsilon$ can go down to the asymptotically achievable precision. Further, to actually find the local maxima of the new criterion, a very simple Iterative Thresholding and K (signed) Means algorithm (ITKM), which has complexity $O(dKN)$ in each iteration, is presented and its local efficiency is demonstrated in several experiments.