Zhengdao Wang

Iowa State University

135Publications

25H-index

5,485Citations

Publications 135

Newest*
*

PA-GD: On the Convergence of Perturbed Alternating Gradient Descent to Second-Order Stationary Points for Structured Nonconvex Optimization*
*

Published on Jun 9, 2019 in ICML (International Conference on Machine Learning)

Estimated H-index: 7

Estimated H-index: 25

Published on May 1, 2019 in ICC (International Conference on Communications)

Estimated H-index: 38

Published on May 1, 2019 in ICASSP (International Conference on Acoustics, Speech, and Signal Processing)

Estimated H-index: 25

Convolutional neural networks have achieved great success in speech, image, and video signal processing tasks in recent years. There have been several attempts to justify the convolutional architecture and to generalize the convolution operation for treatment of other data types such as graphs and manifolds. Based on group representation theory and noncommutative harmonic analysis, it has recently been shown that the so-called group equivariance requirement of a feed-forward neural network neces...

Published on May 1, 2019 in ICASSP (International Conference on Acoustics, Speech, and Signal Processing)

Estimated H-index: 7

Estimated H-index: 26

Estimated H-index: 25

In this paper, we use the perturbed gradient based alternating minimization for solving a class of low-rank matrix factorization problems. Alternating minimization is a simple but popular approach which has been applied to problems in optimization, machine learning, data mining, and signal processing, etc. By leveraging the block structure of the problem, the algorithm updates two blocks of variables in an alternating manner. For the nonconvex optimization problem, it is well-known the alternati...

Superposition coded OFDM transmissions in a downlink cooperative relay network based on statistical channel state information*
*

Estimated H-index: 52

Non-orthogonal multiple access has recently emerged as a promising multiple access technique for future wireless technology. In this study, the authors investigate superposition coding in a downlink cooperative cellular system based on the orthogonal frequency division multiplexing (OFDM) modulation in the presence of frequency selective multipath fading channels, a setup that has not been studied before. Specifically, the base station communicates with two paired mobile users simultaneously via...

Published on Mar 1, 2019

Estimated H-index: 25

The group convolution and representation theory give a strong support for generalized convolutional neural network. The generalized convolutional neural network (G-CNN) has been applied to learning problems and achieved the state-of-art performance. But a theoretical support for details of network architecture design is still lacking. In this work, we first analyze the necessary and sufficient condition for a neural network to be group equivariant when the group acts on the sub-domain of input/o...

Estimated H-index: 1

Estimated H-index: 47

Traditional approaches to secret key establishment based on common randomness have been based on certain restrictive assumptions, such as considering the available common randomness to consist of independent and identically distributed (i.i.d) repetitions of correlated random variables. Unfortunately, the i.i.d assumption does not generally reflect the conditions of real-life scenarios. For this reason, the current paper investigates the key-establishment potential of a more pragmatic model, in ...

Training Optimization and Performance of Single Cell Uplink System With Massive-Antennas Base Station*
*

Estimated H-index: 7

Estimated H-index: 25

We study the performance of uplink transmission in single-cell wireless systems, where all the transmitters have single antennas and the base station has a large number of antennas. We consider both maximum ratio combining and zero-forcing receivers and both small- and large-scale fading channels. We also characterize the achievable total degrees of freedom (DoF) of such a system without assuming channel state information at the receiver. The system DoF turns out to be the same as that of a sing...

Estimated H-index: 7

Estimated H-index: 1

Estimated H-index: 25

We consider a network of pairs of nodes that perform ad hoc simultaneous communications over frequency-selective channels. We assume that the whole frequency band is divided into a number of subbands, and each transmitter can only use one subband. Assuming that the network is geometrically infinite, we use the transmission capacity (TC) as a measure of the network throughput. We consider the problem of allocating nodes to the subbands so that the total TC is maximized, under the constraint of a ...

Estimated H-index: 1

Estimated H-index: 1

Estimated H-index: 14

In wide angle synthetic aperture radar (WASAR), the scattering behavior varies with aspect angle. Subaperture methods can be adopted to accommodate the anisotropic behavior. Modified compressive sensing (MCS) focuses on reconstructing a sparse signal from a limited number of its linear projections when a part of its support is known. It can achieve the accurate recovery with less samples compared with the traditional compressive sensing (CS). Since the support sets of different subaperture image...

12345678910*
*