Mechanism insights into photo-assisted peroxymonosulfate activation on oxygen vacancy-enriched nolanites via an electron transfer regime.

The utilization of photo-assisted persulfate activation for the removal of organic contaminants in water has garnered significant research interest in recent times. However, there remains a lack of clarity regarding specific contributions of light irradiation and catalyst structure in this process. Herein, a photo-assisted peroxymonosulfate (PMS) activation system is designed for the highly efficient degradation of organic contaminants on oxygen vacancy-enriched nolanites (Vo-FVO). Results suggest that the degradation of bisphenol A (BPA) in this system is a nonradical-dominated process via an electron transfer regime, in which VO improves the local electron density and thus facilitates the electron shuttling between BPA and PMS. During BPA degradation, PMS adsorbed at the surface of FVO-180 withdraws electrons near VO and forms FVO-PMS* complexes. Upon light irradiation, photoelectrons effectively restore the electron density around VO, thereby enabling a sustainable electron transfer for the highly efficient degradation of BPA. Overall, this work provides new insights into the mechanism of persulfate activation based on defects engineering in nolanite minerals.

High-precision single-pixel 3D calibration method using pseudo-phase matching.

Compressive sensing makes it possible to explore two-dimensional spatial information using a single-point detector. However, the reconstruction of the three-dimensional (3D) morphology using a single-point sensor is largely limited by the calibration. Here we demonstrate a pseudo-single-pixel camera calibration (PSPC) method using pseudo phase matching in stereo, which can perform 3D calibration of low-resolution images with the help of a high-resolution digital micromirror device (DMD) in the system. In this paper, we use a high-resolution CMOS to pre-image the DMD surface and successfully calibrate the spatial position of a single-point detector and the projector with the support of binocular stereo matching. Our system achieved sub-millimeter reconstructions of spheres, steps, and plaster portraits at low compression ratios with a high-speed digital light projector (DLP) and a highly sensitive single-point detector.

DaisyRec 2.0: Benchmarking Recommendation for Rigorous Evaluation.

Recently, one critical issue looms large in the field of recommender systems - there are no effective benchmarks for rigorous evaluation - which consequently leads to unreproducible evaluation and unfair comparison. We, therefore, conduct studies from the perspectives of practical theory and experiments, aiming at benchmarking recommendation for rigorous evaluation. Regarding the theoretical study, a series of hyper-factors affecting recommendation performance throughout the whole evaluation chain are systematically summarized and analyzed via an exhaustive review on 141 papers published at eight top-tier conferences within 2017-2020. We then classify them into model-independent and model-dependent hyper-factors, and different modes of rigorous evaluation are defined and discussed in-depth accordingly. For the experimental study, we release DaisyRec 2.0 library by integrating these hyper-factors to perform rigorous evaluation, whereby a holistic empirical study is conducted to unveil the impacts of different hyper-factors on recommendation performance. Supported by the theoretical and experimental studies, we finally create benchmarks for rigorous evaluation by proposing standardized procedures and providing performance of ten state-of-the-arts across six evaluation metrics on six datasets as a reference for later study. Overall, our work sheds light on the issues in recommendation evaluation, provides potential solutions for rigorous evaluation, and lays foundation for further investigation.

High-precision nanosecond detection of a gas absorption spectrum based on optical frequency comb time-frequency mapping.

There is an increasing demand for high-precision gas absorption spectroscopy in basic research and industrial applications, such as gas tracking and leak warning. In this Letter, a novel, to the best of our knowledge, high-precision and real-time gas detection method is proposed. A femtosecond optical frequency comb is used as the light source, and a broadening pulse containing a range of oscillation frequencies is formed after passing through a dispersive element and a Mach-Zehnder interferometer. Four absorption lines of H13C14N gas cells are measured at five different concentrations within a single pulse period. A single scan detection time of only 5 ns is obtained along with a coherence averaging accuracy of 0.0055 nm. High-precision and ultrafast detection of the gas absorption spectrum is accomplished while overcoming complexities related to the acquisition system and light source that are encountered in existing methods.

Adversary Agnostic Robust Deep Reinforcement Learning.

Deep reinforcement learning (DRL) policies have been shown to be deceived by perturbations (e.g., random noise or intensional adversarial attacks) on state observations that appear at test time but are unknown during training. To increase the robustness of DRL policies, previous approaches assume that explicit adversarial information can be added into the training process, to achieve generalization ability on these perturbed observations as well. However, such approaches not only make robustness improvement more expensive but may also leave a model prone to other kinds of attacks in the wild. In contrast, we propose an adversary agnostic robust DRL paradigm that does not require learning from predefined adversaries. To this end, we first theoretically show that robustness could indeed be achieved independently of the adversaries based on a policy distillation (PD) setting. Motivated by this finding, we propose a new PD loss with two terms: 1) a prescription gap maximization (PGM) loss aiming to simultaneously maximize the likelihood of the action selected by the teacher policy and the entropy over the remaining actions and 2) a corresponding Jacobian regularization (JR) loss that minimizes the magnitude of gradients with respect to the input state. The theoretical analysis substantiates that our distillation loss guarantees to increase the prescription gap and hence improves the adversarial robustness. Furthermore, experiments on five Atari games firmly verify the superiority of our approach compared to the state-of-the-art baselines.

Generative Multiform Bayesian Optimization.

Many real-world problems, such as airfoil design, involve optimizing a black-box expensive objective function over complex-structured input space (e.g., discrete space or non-Euclidean space). By mapping the complex-structured input space into a latent space of dozens of variables, a two-stage procedure labeled as generative model-based optimization (GMO), in this article, shows promise in solving such problems. However, the latent dimension of GMO is hard to determine, which may trigger the conflicting issue between desirable solution accuracy and convergence rate. To address the above issue, we propose a multiform GMO approach, namely, generative multiform optimization (GMFoO), which conducts optimization over multiple latent spaces simultaneously to complement each other. More specifically, we devise a generative model which promotes a positive correlation between latent spaces to facilitate effective knowledge transfer in GMFoO. And furthermore, by using Bayesian optimization (BO) as the optimizer, we propose two strategies to exchange information between these latent spaces continuously. Experimental results are presented on airfoil and corbel design problems and an area maximization problem as well to demonstrate that our proposed GMFoO converges to better designs on a limited computational budget.

Transfer Kernel Learning for Multi-Source Transfer Gaussian Process Regression.

Multi-source transfer regression is a practical and challenging problem where capturing the diverse relatedness of different domains is the key of adaptive knowledge transfer. In this article, we propose an effective way of explicitly modeling the domain relatedness of each domain pair through transfer kernel learning. Specifically, we first discuss the advantages and disadvantages of existing transfer kernels in handling the multi-source transfer regression problem. To cope with the limitations of the existing transfer kernels, we further propose a novel multi-source transfer kernel kms. The proposed kms assigns a learnable parametric coefficient to model the relatedness of each inter-domain pair, and simultaneously regulates the relatedness of the intra-domain pair to be 1. Moreover, to capture the heterogeneous data characteristics of multiple domains, kms exploits different standard kernels for different domain pairs. We further provide a theorem that not only guarantees the positive semi-definiteness of kms but also conveys a semantic interpretation to the learned domain relatedness. Moreover, the theorem can be easily used in the learning of the corresponding transfer Gaussian process model with kms. Extensive empirical studies show the effectiveness of our proposed method on domain relatedness modelling and transfer performance.

Arbitrary distance measurement without dead zone by chirped pulse spectrally interferometry using a femtosecond optical frequency comb.

We demonstrate an arbitrary distance measurement method by chirped pulse spectrally interferometry (CPSI) using femtosecond optical frequency comb (OFC). In this paper, the chirped fiber Bragg grating (CFBG) is used to investigate the mapping relationship between displacement and the center frequency of the chirped spectral interferogram. We overcome the direction ambiguity of dispersive interferometry (DPI) ranging and expand the range of distance measurement to 18 cm. Besides, we achieve a full range of dead-zone free ranging by introducing a variable optical delay line (VODL). And through principles simulation and experiment, it is demonstrated that the measurement accuracy is 12 µm in comparison with an incremental He-Ne laser interferometer and the minimum Allen deviation is 52 nm at an average time of 1.76 ms. Similarly, in the experiment with long-distance of ∼30m, the accuracy reaches 20 µm, and 2.51 µm repeatability is achieved under harsh environment.

Improvement of Distance Measurement Based on Dispersive Interferometry Using Femtosecond Optical Frequency Comb.

Since the dispersive interferometry (DPI) based on optical frequency combs (OFCs) was proposed, it has been widely used in absolute distance measurements with long-distance and high precision. However, it has a serious problem for the traditional DPI based on the mode-locked OFC. The error of measurements caused by using the fast Fourier transform (FFT) algorithm to process signals cannot be overcome, which is due to the non-uniform sampling intervals in the frequency domain of spectrometers. Therefore, in this paper, we propose a new mathematical model with a simple form of OFC to simulate and analyze various properties of the OFC and the principle of DPI. Moreover, we carry out an experimental verification, in which we adopt the Lomb-Scargle algorithm to improve the accuracy of measurements of DPI. The results show that the Lomb-Scargle algorithm can effectively reduce the error caused by the resolution, and the error of absolute distance measurement is less than 12 µm in the distance of 70 m based on the mode-locked OFC.

Nonequal arm surface measurement of femtosecond optical frequency combs using the Savitzky-Golay filtering algorithm.

In precision machining, the surface geometry of a device is one of the important parameters that directly affects the device performance. This paper proposes nonequal arm surface measurement of femtosecond optical frequency combs (OFCs) using the Savitzky-Golay filtering algorithm, which uses the high spatial coherence of OFCs to realize high-precision, nonequal surface measurements. The Savitzky-Golay filtering algorithm and a high-order polynomial envelope fitting algorithm are used to smooth and denoise the interference signals to improve signal quality and measurement accuracy. The experiments are carried out under the condition of nonequal arms, and the results show that the repeatability is 28.6 nm for 20 consecutive measurements on the step surface of a 0.5 mm gauge block. The frosted glass surface is measured 20 times, and the measurement repeatability at the center position is 89.6 nm, which verified the system capability of nonequal arm high-precision measurement under different reflective surfaces.

Frame-Correlation Transfers Trigger Economical Attacks on Deep Reinforcement Learning Policies.

Adversarial attack can be deemed as a necessary prerequisite evaluation procedure before the deployment of any reinforcement learning (RL) policy. Most existing approaches for generating adversarial attacks are gradient based and are extensive, viz., perturbing every pixel of every frame. In contrast, recent advances show that gradient-free selective perturbations (i.e., attacking only selected pixels and frames) could be a more realistic adversary. However, these attacks treat every frame in isolation, ignoring the relationship between neighboring states of a Markov decision process; thus resulting in high computational complexity that tends to limit their real-world plausibility due to the tight time constraint in RL. Given the above, this article showcases the first study of how transferability across frames could be exploited for boosting the creation of minimal yet powerful attacks in image-based RL. To this end, we introduce three types of frame-correlation transfers (FCTs) (i.e., anterior case transfer, random projection-based transfer, and principal components-based transfer) with varying degrees of computational complexity in generating adversaries via a genetic algorithm. We empirically demonstrate the tradeoff between the complexity and potency of the transfer mechanism by exploring four fully trained state-of-the-art policies on six Atari games. Our FCTs dramatically speed up the attack generation compared to existing methods, often reducing the computation time required to nearly zero; thus, shedding light on the real threat of real-time attacks in RL.

Subdomain Adaptation With Manifolds Discrepancy Alignment.

Reducing domain divergence is a key step in transfer learning. Existing works focus on the minimization of global domain divergence. However, two domains may consist of several shared subdomains, and differ from each other in each subdomain. In this article, we take the local divergence of subdomains into account in transfer. Specifically, we propose to use the low-dimensional manifold to represent the subdomain, and align the local data distribution discrepancy in each manifold across domains. A manifold maximum mean discrepancy (M3D) is developed to measure the local distribution discrepancy in each manifold. We then propose a general framework, called transfer with manifolds discrepancy alignment (TMDA), to couple the discovery of data manifolds with the minimization of M3D. We instantiate TMDA in the subspace learning case considering both the linear and nonlinear mappings. We also instantiate TMDA in the deep learning framework. Experimental studies show that TMDA is a promising method for various transfer learning tasks.

Precision and repeatability improvement in frequency-modulated continuous-wave velocity measurement based on the splitting of beat frequency signals.

The basic principle of frequency-modulated continuous-wave lidars is to measure the velocity of a moving object through the Doppler frequency shift phenomenon. However, the vibration generated by the moving object will cause the spectrum to broaden and the precision and repeatability of speed measurement to decrease. In this paper, we propose a speed measurement method based on H13C14N gas cell absorption peak splitting the sweep signal of a large bandwidth triangular wave modulated frequency laser. This method obtains the speed of a continuously moving target by re-splicing an accurately-split frequency sweep signal, which effectively solves the problem of simultaneous processing of excessive amounts of data when measuring the speed of a continuously moving target. At the same time, the H13C14N gas cell absorbs the spectra of specific wavelengths, which reduces the phase delay of the beat signal corresponding to the up- and down-scanning, thus reducing the signal spectrum broadening caused by frequency deviation, and improving the speed measurement resolution and range effectively. The experimental results show that for speeds of up to 30mm/s, the mean error was less than 23µm/s and the mean standard deviation was less than 61µm/s.

Fast HDR image generation method from a single snapshot image based on frequency division multiplexing technology.

Traditional high dynamic range (HDR) image generation algorithms such as multi-exposure fusion need to capture multiple images for algorithm fusion, which is not only slow but also occupies a lot of storage space, which limits the application of multi-exposure fusion technology. In this paper, the frequency division multiplexing method is used to separate the sub-images with different exposure values from a single snapshot image successfully. The resolution of HDR images generated by this method is almost the same as that of the traditional multiple exposure methods, the storage space is greatly reduced and the imaging speed is improved.

Nonlinear calibration of frequency modulated continuous wave LIDAR based on a microresonator soliton comb.

Traditional frequency modulated continuous wave (FMCW) LIDAR ranging is based on heterodyne detection, calculating unknown distance by extracting the frequency of the interference signal, while the main error source is frequency modulation (FM) nonlinearity. In this paper, a ranging system based on a microresonator soliton comb is demonstrated to correct the nonlinearity by sampling the ranging signals at equal frequency intervals, producing a ranging error lower than 20 µm, while at the range of 2 m. Advantages of fast data acquisition, light computation requirements, and a simple optical path, without long optical fiber, give this method a high practical value in precision manufacturing.

Performance Evaluation and Compensation Method of Trigger Probes in Measurement Based on the Abbé Principle.

Trigger probes are widely used in precision manufacturing industries such as coordinate measuring machines (CMM) and high-end computer numerical control(CNC) machine tools for quality control. Their performance and accuracy often determine the measurement results and the quality of the product manufacturing. However, because there is no accurate measurement of the trigger force in different directions of the probe, and no special measuring device to calibrate the characteristic parameters of the probe in traditional measurement methods, it is impossible to exactly compensate for the measurement error caused by the trigger force of the probe in the measurement process. The accuracy of the measurement of the equipment can be improved by abiding by the Abbé principle. Thus, in order to better evaluate the performance parameters of the probe and realize the accurate compensation for its errors, this paper presents a method which can directly measure the performance parameters of the trigger probe based on the Abbé measurement principle, expounds the measurement principle, the establishment of the mathematical model, and the calibration system, and finishes with an experimental verification and measurement uncertainty analysis. The experimental results show that this method can obtain the exact calibration errors of the performance parameters of the trigger probe intuitively, realize the compensation for the errors of the probe in the measurement process, and effectively improve the measurement accuracy.

Microdeformation of RBCs under oxidative stress measured by digital holographic microscopy and optical tweezers.

This paper utilized digital holographic microscopy and optical tweezers to study microdeformation of red blood cells (RBCs) dynamically under oxidative stress. RBCs attached with microbeads were stretched by dual optical tweezers to generate microdeformation. Morphology of RBCs under manipulation were recorded dynamically and recovered by off-axis digital holographic microscopy method. RBCs treated with H2O2 at different concentrations were measured to investigate the mechanical properties under oxidative stress. Use of optical tweezers and off-axis digital holographic microscopy enhanced measuring accuracy compared with the traditional method. Microdeformation of RBCs is also more consistent with the physiological situation. This proposal is meaningful for clinical applications and basic analysis of Parkinson's disease research.

Laser interference-based technique for dynamic measurement of single cell deformation manipulated by optical tweezers.

A laser interference-based method was proposed to measure the deformation response of cell manipulated by optical tweezers. This method was implemented experimentally by integrating a laser illuminating system and optical tweezers with an inverted microscope. Interference fringes generated by the transmitted and reflected lights were recorded by a complementary metal oxide semiconductor camera. From the acquired images, cell height was calculated and cell morphology was constructed. To further validate this method, the morphological analyses of HeLa cells were performed in static state and during detachment process. Subsequently, the dynamic deformation responses of red blood cells were measured during manipulation with optical tweezers. Collectively, this laser interference-based method precludes the requirement of complex optical alignment, allows easy integration with optical tweezers, and enables dynamic measurement of cell deformation response by using a conventional inverted microscope.

Micron-precision measurement using a combined frequency-modulated continuous wave ladar autofocusing system at 60 meters standoff distance.

In this paper, we propose a novel combined frequency-modulated continuous wave (FMCW) ladar autofocusing system and a fast compensation method for dispersion mismatch, which could allow high-precision ranging to be performed at a long distance. By using the dual-beam laser autofocusing system based on a liquid lens, this system can quickly complete a measurement with high-precision. The experimental results showed that the precision was below 126 µm in a range up to 60m, corresponding to a relative precision of 2.1 × 10 -6, compared to a reference interferometer.

Nonlinear error correction for FMCW ladar by the amplitude modulation method.

FMCW ladar is a kind of absolute distance measurement technology with high spatial resolution. However, the advantage of high spatial resolution is significantly covered up by the non-linearity of laser frequency sweep. One of the typical approaches for the nonlinearity is resample technology, which has residual phase error from the sample time delay mismatch between the clock signal and the measurement signal. We have proposed and demonstrated a novel amplitude modulation method for correcting the nonlinear error of FMCW technology. The optical structure of the method is comprised of two tandem fiber interferometers. The first interferometer is used to produce a carrier signal and the second one is used to load the range information on the amplitude of the carrier signal. In the end, the experimental result verifies that the nonlinear error can be suppressed effectively, the phase error from the mismatch has been eliminated observably, and the range resolution can be notably improved to 69µm; the stability is 2.9µm and the measurement precision is 4.3µm.