Deep-Energy: Unsupervised Training of Deep Neural Networks

TL;DR

Deep Energy introduces an unsupervised training method for deep neural networks that leverages task-specific energy functions, enabling label inference directly from real-world data without manual annotations or synthetic data, demonstrated across multiple tasks.

Contribution

The paper presents a novel unsupervised training approach using energy functions, allowing DNNs to learn from real inputs alone without labeled data or simulation, broadening applicability.

Findings

Network solutions often outperform direct energy minimization.

Method is effective across diverse tasks like segmentation, matting, and dehazing.

Approach exhibits strong generality and wide applicability.

Abstract

The success of deep learning has been due, in no small part, to the availability of large annotated datasets. Thus, a major bottleneck in current learning pipelines is the time-consuming human annotation of data. In scenarios where such input-output pairs cannot be collected, simulation is often used instead, leading to a domain-shift between synthesized and real-world data. This work offers an unsupervised alternative that relies on the availability of task-specific energy functions, replacing the generic supervised loss. Such energy functions are assumed to lead to the desired label as their minimizer given the input. The proposed approach, termed "Deep Energy", trains a Deep Neural Network (DNN) to approximate this minimization for any chosen input. Once trained, a simple and fast feed-forward computation provides the inferred label. This approach allows us to perform unsupervised training of DNNs with real-world inputs only, and without the need for manually-annotated labels, nor synthetically created data. "Deep Energy" is demonstrated in this paper on three different tasks -- seeded segmentation, image matting and single image dehazing -- exposing its generality and wide applicability. Our experiments show that the solution provided by the network is often much better in quality than the one obtained by a direct minimization of the energy function, suggesting an added regularization property in our scheme.