On the existence of optimal multi-valued decoders and their accuracy bounds for undersampled inverse problems

Abstract: Undersampled inverse problems occur everywhere in the sciences including medical imaging, radar, astronomy etc., yielding underdetermined linear or non-linear reconstruction problems. There are now a myriad of techniques to design decoders that can tackle such problems, ranging from optimization based approaches, such as compressed sensing, to deep learning (DL), and variants in between the two techniques. The variety of methods begs for a unifying approach to determine the existence of optimal decoders and fundamental accuracy bounds, in order to facilitate a theoretical and empirical understanding of the performance of existing and future methods. Such a theory must allow for both single-valued and multi-valued decoders, as underdetermined inverse problems typically have multiple solutions. Indeed, multi-valued decoders arise due to non-uniqueness of minimizers in optimisation problems, such as in compressed sensing, and for DL based decoders in generative adversarial models, such as diffusion models and ensemble models. In this work we provide a framework for assessing the lowest possible reconstruction accuracy in terms of worst- and average-case errors. The universal bounds bounds only depend on the measurement model $F$, the model class $\mathcal{M}_1 \subseteq \mathcal{X}$ and the noise model $\mathcal{E}$. For linear $F$ these bounds depend on its kernel, and in the non-linear case the concept of kernel is generalized for undersampled settings. Additionally, we provide multi-valued variational solutions that obtain the lowest possible reconstruction error.