Recurrent Inference Machines for Solving Inverse Problems

TL;DR

Recurrent Inference Machines (RIM) are a novel learning framework that trains RNNs to learn inference algorithms directly from data, enabling flexible, domain-agnostic solutions for inverse problems like image denoising and super-resolution.

Contribution

The paper introduces RIM, a new approach that trains RNNs to learn inference algorithms, removing the need for domain-specific algorithm design and achieving state-of-the-art results.

Findings

Achieved state-of-the-art performance in image denoising.

Outperformed existing methods in super-resolution tasks.

Demonstrated superior cross-task generalization.

Abstract

Much of the recent research on solving iterative inference problems focuses on moving away from hand-chosen inference algorithms and towards learned inference. In the latter, the inference process is unrolled in time and interpreted as a recurrent neural network (RNN) which allows for joint learning of model and inference parameters with back-propagation through time. In this framework, the RNN architecture is directly derived from a hand-chosen inference algorithm, effectively limiting its capabilities. We propose a learning framework, called Recurrent Inference Machines (RIM), in which we turn algorithm construction the other way round: Given data and a task, train an RNN to learn an inference algorithm. Because RNNs are Turing complete [1, 2] they are capable to implement any inference algorithm. The framework allows for an abstraction which removes the need for domain knowledge. We demonstrate in several image restoration experiments that this abstraction is effective, allowing us to achieve state-of-the-art performance on image denoising and super-resolution tasks and superior across-task generalization.