Learning to Optimize: Training Deep Neural Networks for Interference Management

By Haoran Sun, Xiangyi Chen, Qingjiang Shi, Mingyi Hong, Xiao Fu, N. Sidiropoulos. 2017

725 citations recorded by the source corpus.

Numerical optimization has played a central role in addressing key signal processing (SP) problems. Highly effective methods have been developed for a large variety of SP applications such as communications, radar, filter design, and speech and image analytics, just to name a few. However, optimization algorithms often entail considerable complexity, which creates a serious gap between theoretical design/analysis and real-time processing. In this paper, we aim at providing a new learning-based perspective to address this challenging issue. The key idea is to treat the input and output of an SP algorithm as an unknown nonlinear mapping and use a deep neural network (DNN) to approximate it. If the nonlinear mapping can be learned accurately by a DNN of moderate size, then SP tasks can be performed effectively—since passing the input through a DNN only requires a small number of simple operations. In our paper, we first identify a class of optimization algorithms that can be accurately approximated by a fully connected DNN. Second, to demonstrate the effectiveness of the proposed approach, we apply it to approximate a popular interference management algorithm, namely, the WMMSE algorithm. Extensive experiments using both synthetically generated wireless channel data and real DSL channel data have been conducted. It is shown that, in practice, only a small network is sufficient to obtain high approximation accuracy, and DNNs can achieve orders of magnitude speedup in computational time compared to the state-of-the-art interference management algorithm.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Electrical and Electronic Engineering

Citations in the corpus

Deep Learning for Distributed Optimization: Applications to Wireless Resource Management (2019, methodology): “Compared to a supervised learning technique [7], the proposed DL framework is based on an unsupervised learning strategy that does not require training labels, i.e., the predetermined optimal solutions.”
Edge Learning for B5G Networks With Distributed Signal Processing: Semantic Communication, Edge Computing, and Wireless Sensing (2022, positive): “It has been applied to power allocation [177], precoding design [43], [178], [179], and other end-to-end designs in communication systems.”
Semantic Communication Systems for Speech Transmission (2021, background): “Particularly, DL has shown its great potentials to solve the existing technical problems in both physical layer communications [3]–[5] and wireless resource allocations [6], [7].”
A Deep Learning Framework for Optimization of MISO Downlink Beamforming (2019, methodology): “The WMMSE algorithm is a good choice to find the locally optimal solutions [9], [10], and such an iterative algorithm ensures the continuity of the mapping from the channel to the solution, and can be learned by a neural network [27], [30].”
Application of Machine Learning in Wireless Networks: Key Techniques and Open Issues (2018, methodology): “Specifically, for power allocation, the aim of authors in [15], [55], [56] is to utility neural networks to approximate high complexity power allocation algorithms, such as the genetic algorithm and WMMSE algorithm.”
Deep Learning Enabled Semantic Communication Systems (2020, background): “Recent advancements on deep learning (DL) based natural language processing (NLP) and communication systems inspire us to investigate semantic communication to realize the second level communications as aforementioned [10]–[15].”
Holistic Network Virtualization and Pervasive Network Intelligence for 6G (2023, methodology): “• Management-Oriented AI Applications - In these applications, AI is used as a tool for network management, such as transmission power allocation in cellular networks [139] and resource reservation in network slices [19].”
Graph Neural Networks for Scalable Radio Resource Management: Architecture Design and Theoretical Analysis (2020, methodology): “For example, it was shown in [7] that the performance gap to the WMMSE algorithm is 2% when K = 10 and it becomes 12% when K = 30.”
Edge Artificial Intelligence for 6G: Vision, Enabling Technologies, and Applications (2021, methodology): “A multi-layer perceptron was adopted in [265] to directly learn the mapping from the problem instance to the output solution generated by the weighted minimum mean square error (WMMSE) algorithm for nonconvex precoding design [266].”
Reconfigurable Intelligent Surface Assisted Multiuser MISO Systems Exploiting Deep Reinforcement Learning (2020, neutral): “Similarly, a detailed explanation and pseudo code of the WMMSE algorithm is given in [43] Algorithm 1.”
Multi-Agent Deep Reinforcement Learning for Dynamic Power Allocation in Wireless Networks (2018, methodology): “As the network gets larger, the total number of DNN’s input and output ports also increases, which raises questions on the scalability of the centralized solution of [9].”
A Tutorial on Ultrareliable and Low-Latency Communications in 6G: Integrating Domain Knowledge Into Deep Learning (2021, methodology): “More recently, the results in [221] showed that to achieve a target accuracy, the size of network scales with the approximation accuracy ε NN according to O (ln(1 /ε NN )) .”
Machine Learning in the Air (2019, background): “These algorithms are too complex for practical implementation, but a key idea advocated in [18], [53] is that we can take this complexity ofﬂine by training a DNN to mimic the input-output behavior of the WMMSE algorithm.”
Constrained Deep Learning for Wireless Resource Management (2019, methodology): “, the WMMSE solution in [3], during the training of the DNNs.”