Search

With the explosive developments of deep learning, learning–based computer–generated holography (CGH) has become an effective way to achieve real–time and high–quality holographic displays. Plentiful learning–based methods with various deep neural networks (DNNs) have been proposed. In this paper, we focus on the rapid progress of learning–based CGH in recent years. The generation principles and algorithms of CGH are introduced. The DNN structures frequently used in CGH are compared, including U–Net, ResNet, and GAN. We review the developments and discuss the outlook of the learning–based CGH. [ABSTRACT FROM AUTHOR]

Three-term conjugate gradient methods have attracted much attention for large-scale unconstrained problems in recent years, since they have attractive practical factors such as simple computation, low memory requirement, better descent property and strong global convergence property. In this paper, a hybrid three-term conjugate gradient algorithm is proposed and it owns a sufficient descent property, independent of any line search technique. Under some mild conditions, the proposed method is globally convergent for uniformly convex objective functions. Meanwhile, by using the modified secant equation, the proposed method is also global convergence without convexity assumption on the objective function. Numerical results also indicate that the proposed algorithm is more efficient and reliable than the other methods for the testing problems. [ABSTRACT FROM AUTHOR]

In this paper, an adaptive proximal bundle method is proposed for a class of nonconvex and nonsmooth composite problems with inexact information. The composite problems are the sum of a finite convex function with inexact information and a nonconvex function. For the nonconvex function, we design the convexification technique and ensure the linearization errors of its augment function to be nonnegative. Then, the sum of the convex function and the augment function is regarded as an approximate function to the primal problem. For the approximate function, we adopt a disaggregate strategy and regard the sum of cutting plane models of the convex function and the augment function as a cutting plane model for the approximate function. Then, we give the adaptive nonconvex proximal bundle method. Meanwhile, for the convex function with inexact information, we utilize the noise management strategy and update the proximal parameter to reduce the influence of inexact information. The method can obtain an approximate solution. Two polynomial functions and six DC problems are referred to in the numerical experiment. The preliminary numerical results show that our algorithm is effective and reliable. [ABSTRACT FROM AUTHOR]

To reduce the intense terahertz (THz) wave absorption of water and increase the signal-to-noise ratio, the THz spectroscopy detection of biomolecules usually operates using the nanofluidic channel technologies in practice. The effects of confinement due to the existence of nanofluidic channels on the conformation and dynamics of biomolecules are well known. However, studies of confinement effects on the THz spectra of biomolecules are still not clear. In this work, extensive all-atom molecular dynamics simulations are performed to investigate the THz spectra of the glycine oligopeptide solutions in free and confined environments. THz spectra of the oligopeptide solutions confined in carbon nanotubes (CNTs) with different radii are calculated and compared. Results indicate that with the increase of the degree of confinement (the reverse of the radius of CNT), the THz absorption coefficient decreases monotonically. By analyzing the diffusion coefficient and dielectric relaxation dynamics, the hydrogen bond life, and the vibration density of the state of the water molecules in free solution and in CNTs, we conclude that the confinement effects on the THz spectra of biomolecule solutions are mainly to slow down the dynamics of water molecules and hence to reduce the THz absorption of the whole solution in confined environments. [ABSTRACT FROM AUTHOR]

Terahertz signature detection of biological samples in aqueous solution remains a great challenge due to the strong terahertz absorption of water. Here we propose a new preparation process for fabricating a microfluidic chip and use it as an effective sensor to probe the terahertz absorption signatures of microcystin aptamer (a linear single-stranded DNA with 60 nucleotides) dissolved in TE buffer with different concentrations. The microfluidic chip made of silicon includes thousands of 2.4 μm × 2.4 μm square-cross-section channels. One repeatable terahertz absorption signature is detected and recognized around 830 GHz, fitted to a Lorentz oscillator. This signature is theorized to originate from the bending of hydrogen bonds formed between adjacent hydrated DNA bases surrounded by water molecules. Furthermore, the low-lying vibrational modes are also investigated by molecular dynamics simulations which suggest that strong resonant oscillations are highly probable in the 815–830 GHz frequency band. [ABSTRACT FROM AUTHOR]