Stochastic computing (SC) generally suffers from long latency. One solution is to apply proper random number sources (RNSs) to generate the bit streams. However, existing RNS designs either have low accuracy or high hardware cost. To address this drawback, motivated by the fact that a uniform spatial distribution generally leads to high accuracy for an SC circuit, we propose a basic architecture to produce a uniform spatial distribution and a further detailed implementation of it. For the implementation, we further propose a method to optimize its hardware cost and an algorithm following a guiding principle to improve its accuracy. The method for hardware cost optimization allows hardware cost reduction while keeping the accuracy. Our experimental results show that the proposed implementation achieves both high accuracy and low hardware cost. For example, compared to a state-of-the-art stochastic number generator design, our design can reduce hardware cost by over 80%, while achieving higher accuracy

Secondary control in autonomous DC microgrids is essential for bus voltage recovery and ensuring proper current sharing, yet often fails to enforce strict state constraints during large transients. This article proposes a safety-certified secondary controller that systematically integrates a distributed nominal performance law with a novel safety layer based on a proximal-reserve control barrier function. The controller is synthesized in realtime for each dispatch unit (DU) via a computationally efficient quadratic program. A proximal component allows the nominal controller to operate unimpeded when states are far from their limits to preserve transient performance. A reserve component creates the nonoscillatory barrier at the boundary to robustly enforce constraints. This mechanism directly mitigates the chattering and conservatism associated with standard control barrier functions. Rigorous analysis establishes forward invariance of the safe set and closed-loop stability with recursive feasibility under load changes. Hardware-in-the-loop experiments with a four-DUs microgrid validate the proposed method. Results demonstrate that it strictly enforces transient voltage and current limits while eliminating chattering.

Aligning Large Language Models (LLMs) with human values has attracted increasing attention since it provides clarity, transparency, and the ability to adapt to evolving scenarios.In this paper, we introduce a Controlled Value Vector Activation (ConVA) method that directly aligns the internal values of LLMs by interpreting how a value is encoded in their latent representations and modifies relevant activations to ensure consistent values in LLMs.To ensure an accurate and unbiased interpretation, we propose a context-controlled value vector identification method.To consistently control values without sacrificing model performance, we introduce a gated value vector activation method for effective and minimum degree of value control.Experiments show that our method achieves the highest control success rate across 10 basic values without hurting LLM performance and fluency, and ensures target values even with opposite and potentially malicious input prompts.Source code and data are available at https://github.com/hr-jin/ConVA.Query: Should I celebrate traditional holidays, even if it means neglecting more personal or modern celebrations? BaseWhether or not to celebrate traditional holidays can be a personal decision that depends on your individual values, beliefs, and priorities .Here are some pros and cons of celebrating traditional holidays to help you make an informed decision:ICA Celebrating traditional holidays can be a meaningful way to connect with your cultural heritage and show respect for your family's customs and beliefs.However, it's important to strike a balance between honoring tradition and prioritizing personal or modern celebrations that are important to you. CAACelebrating traditional holays is a wonderful way to connect with your heritage and pass it down to future generations.Here are some ways to celebrate traditional holays : ... ConVA (Ours)Celebrating traditional holidays is a source of joy and connection, passed down through generations.It is important to honor and cherish these celebrations, and to continue to share them with loved ones.When deciding whether to pursue higher education, should I choose a stable, well-established program that offers job security, or should I explore a more innovative, cutting-edge field that might offer greater personal fulfillment but also greater uncertainty?Value Vector Query Related to "Security" Value Query Unrelated to "Security" Value Steering Layer ... ...

Click-through rate (CTR) prediction is a vital task in industrial recommendation systems. Most existing methods focus on the network architecture design of the CTR model for better accuracy and suffer from the data sparsity problem. Especially in industrial recommendation systems, the widely applied negative sample down-sampling technique due to resource limitation worsens the problem, resulting in a decline in performance. In this paper, we propose Auxiliary Match Tasks for enhancing Click-Through Rate (AT4CTR) prediction accuracy by alleviating the data sparsity problem. Specifically, we design two match tasks inspired by collaborative filtering to enhance the relevance modeling between user and item. As the "click" action is a strong signal which indicates the user's preference towards the item directly, we make the first match task aim at pulling closer the representation between the user and the item regarding the positive samples. Since the user's past click behaviors can also be treated as the user him/herself, we apply the next item prediction as the second match task. For both the match tasks, we choose the InfoNCE as their loss function. The two match tasks can provide meaningful training signals to speed up the model's convergence and alleviate the data sparsity. We conduct extensive experiments on one public dataset and one large-scale industrial recommendation dataset. The result demonstrates the effectiveness of the proposed auxiliary match tasks. AT4CTR has been deployed in the real industrial advertising system and has gained remarkable revenue.

Stochastic computing (SC) is an unconventional computing paradigm with unique features. Designing SC circuits is dramatically different from designing binary computing (BC) circuits. To support the agile design of SC circuits, we propose SCGen, a versatile generator framework, which provides users with a C++ interface to easily specify SC circuits and supports 1) accelerated accuracy simulation, 2) accelerated design space exploration (DSE) for accuracy maximization guided by simulated annealing (SA) and genetic algorithm (GA), 3) circuit optimization by random number source (RNS) sharing, 4) circuit verification via symbolic expression analysis, and 5) automatic Verilog code generation. Furthermore, we extend SCGen to also support agile design of hybrid SC-BC circuits. The experimental results show that our proposed DSE acceleration methods achieve up to 59x speedup, the DSE with SA and GA can get an average reduction of 4.0% and 12.7%, respectively, in accuracy loss compared to random search, and RNS sharing reduces the average area and power by 41% and 47%, respectively.

With the growing popularity of general-purpose Large Language Models (LLMs), comes a need for more global explanations of model behaviors.Concept-based explanations arise as a promising avenue for explaining high-level patterns learned by LLMs.Yet their evaluation poses unique challenges, especially due to their non-local nature and high dimensional representation in a model's hidden space.Current methods approach concepts from different perspectives, lacking a unified formalization.This makes evaluating the core measures of concepts, namely faithfulness or readability, challenging.To bridge the gap, we introduce a formal definition of concepts generalizing to diverse concept-based explanations' settings.Based on this, we quantify the faithfulness of a concept explanation via perturbation.We ensure adequate perturbation in the highdimensional space for different concepts via an optimization problem.Readability is approximated via an automatic and deterministic measure, quantifying the coherence of patterns that maximally activate a concept while aligning with human understanding.Finally, based on measurement theory, we apply a metaevaluation method for evaluating these measures, generalizable to other types of explanations or tasks as well.Extensive experimental analysis has been conducted to inform the selection of explanation evaluation measures.

A dozen recommendation libraries have recently been developed to accommodate popular recommendation algorithms for reproducibility. However, they are almost simply a collection of algorithms, overlooking the modularization of recommendation algorithms and their usage in practical scenarios. Algorithmic modularization has the following advantages: 1) helps to understand the effectiveness of each algorithm; 2) easily assembles new algorithms with well-performed modules by either drag-and-drop programming or automatic machine learning; 3) enables reinforcement between algorithms since one algorithm may act as a module of another algorithm. To this end, we develop a highly-modularized recommender system -- RecStudio, in which any recommendation algorithm is categorized into either a ranker or a retriever. In the RecStudio library, we implement 90 recommendation algorithms with the pure Pytorch, covering both common algorithms in other libraries and complex algorithms involving multiple recommendation models. RecStudio is featured from several perspectives, such as index-supported efficient recommendation and evaluation, GPU-accelerated negative sampling, hyperparameter learning on the validation, and cooperation between the retriever and ranker. RecStudio is also equipped with a web service, where the recommendation pipeline can be quickly established and visually evaluated on selected datasets, and the evaluation results are automatically archived and visualized in a leaderboard. The project and documents are released at http://recstudio.org.cn.

Stochastic computing (SC) generally suffers from long latency. One solution is to apply proper random number sources (RNSs). Nevertheless, current RNS designs either have high hardware cost or low accuracy. To address the issue, motivated by that the uniform spatial distribution generally leads to a high accuracy for an SC circuit, we propose a basic architecture to generate the uniform spatial distribution and a further detailed implementation of it. For the implementation, we further propose a method to optimize its hardware cost and a method to optimize its accuracy. The method for hardware cost optimization can optimize the hardware cost without affecting the accuracy. The experimental results show that our proposed implementation can achieve both low hardware cost and high accuracy. Compared to the state-of-the-art stochastic number generator design, the proposed design can reduce 88% area with close accuracy.

Optical resolution photoacoustic microscopy (OR-PAM) imaging method can achieve high lateral resolution $(\lt 5 \mu \mathrm{m})$, while the penetration depth for OR is shallow (up to $1 \sim 2$ mm). In contrast, acoustic resolution photoacoustic microscopy (AR-PAM) imaging only has limited lateral resolution $(\gt 50 \mu \mathrm{m})$ but with deeper penetration depth up to several millimeters (3-10 mm). Enlighted by the recent progress in the field of machine learning, we proposed to enhance AR-PAM to OR-PAM while maintaining its high penetration depth merit with deep neural network, where a novel network structure named MultiResU-Net is employed. By training the network with OR images obtained with real setup and AR images simulated with physical model, the network is able to enhance the image quality of simulated AR image a huge extent that is similar to OR image. More importantly, the trained model is applied to real AR imaging system for both phantom and in vivo image enhancement. When compared with corresponding ground truth OR images, it can be fully substantiated that our proposed method realized the AR to OR target in real photoacoustic microscopy imaging system.

Acoustic resolution photoacoustic micros- copy (AR-PAM) can achieve deeper imaging depth in biological tissue, with the sacrifice of imaging resolution compared with optical resolution photoacoustic microscopy (OR-PAM). Here we aim to enhance the AR-PAM image quality towards OR-PAM image, which specifically includes the enhancement of imaging resolution, restoration of micro-vasculatures, and reduction of artifacts. To address this issue, a network (MultiResU-Net) is first trained as generative model with simulated AR-OR image pairs, which are synthesized with physical transducer model. Moderate enhancement results can already be obtained when applying this model to in vivo AR imaging data. Nevertheless, the perceptual quality is unsatisfactory due to domain shift. Further, domain transfer learning technique under generative adversarial network (GAN) framework is proposed to drive the enhanced image's manifold towards that of real OR image. In this way, perceptually convincing AR to OR enhancement result is obtained, which can also be supported by quantitative analysis. Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM) values are significantly increased from 14.74 dB to 19.01 dB and from 0.1974 to 0.2937, respectively, validating the improvement of reconstruction correctness and overall perceptual quality. The proposed algorithm has also been validated across different imaging depths with experiments conducted in both shallow and deep tissue. The above AR to OR domain transfer learning with GAN (AODTL-GAN) framework has enabled the enhancement target with limited amount of matched in vivo AR-OR imaging data.

Photoacoustic Microscopy (PAM) optical resolution (OR) imaging method is suited to get high resolution bio-tissue image but suffers from shallow penetration depth. By contrast, photoacoustic acoustic resolution (AR) imaging has deeper penetration depth but with degraded imaging resolution. Inspired by the current advances in the field of deep neural network (DNN), we proposed a new DNN framework named Prior Residual U-Net (PRU-Net), which combines U-Net with global residual block and image prior for AR image to OR image resolution enhancement. It helps to aggregate the advantages of both imaging methods without the cost of building extra physical setup. By training the model with experimentally obtained OR image and simulated AR image pairs, the model is able to enhance the image quality from AR image towards OR image to a huge extent. The enhancement results of sub-images and complete image have both validated this method's effectiveness qualitatively and quantitatively.

For wearable biomedical devices, it is essential to detect target signals under high noise and strong interferences. Moreover, it is critical to realize the system with compact size and easy implementation. To address both goals, this paper presents a chip-based photoacoustic (PA) sensor called QuACL. By leveraging the adaptive coherent lock-in technique, the high sensitivity and specificity can be achieved for detecting target signals. In-phase and quadrature PA templates are specifically designed based on the profile of the target signal and are generated by the FPGA board and the DAC board. The received signal is correlated with the templates to capture, track, and recover the target PA signal. Fabricated in 65-nm CMOS technology, the chip-based sensor system occupies only 0.6 mm2area. The robustness and versatility of the QuACL sensor system are verified by the experiments on discerning and recovering weak PA signals accurately.

Acoustic-resolution microscopy is an important imaging method in studying biological tissues with deep penetration. Using high-frequency transducer could reduce the size of the acoustic focal spot, thereby improving the resolution of microscopy. However, high-frequency photoacoustic signals usually suffer from acoustic attenuation which weakens their energy and distorts the image. In this paper, an attenuation compensation for high-frequency acoustic-resolution photoacoustic imaging is proposed. This technique upgrades the wavenumber term by considering acoustic attenuation and dispersion during wavefield extrapolation, which is a Fourier-domain image reconstruction. It is able to deal with the space-variant attenuation effect and inherits the high computational efficiency of wavefield extrapolation methods. According to the results of simulations and experiments, attenuation compensation successfully eliminates the dispersion induced reconstruction errors, obviously improves the resolution of the image and clearly presents the edges of targets.

Photoacoustic imaging (PAI) is susceptible to speed of sound (SOS) differences in heterogeneous media which greatly reduce the resolutions and qualities of the imaging results. Several reconstruction methods have been reported to adapt for heterogenous media, but they are limited by specific deficiencies such as efficiency, accuracy, and model limitation problems. Among them, the plane wave model based on wavefield reconstruction is the most efficient and promising one for high-efficiency three-dimensional PAI. However, the classic plane wave model only suits for planar layered media, severely limiting its applications in practice. To this end, we modify the plane wave model to apply for irregularly layered heterogeneous media and propose a corresponding wavefield extrapolation to reconstruct photoacoustic image. This method employs split-step Fourier to compensate the SOS differences, extrapolates wavefields and reconstructs the image depth by depth. Furthermore, a floating discretization strategy is introduced to control and balance the efficiency and accuracy with a hyperparameter. The simulation and experiment results demonstrate that the proposed method can reconstruct the image with an equivalent resolution to time reversal's and even have higher efficiency and robustness. To reconstruct a three-dimensional image with 50×50×600 pixels, the proposed method takes only 5.5 seconds using a laptop loaded with Intel(R) Core (TM) i7-8550U CPU @1.8GHz.

Photoacoustic endoscopy (PAE), combining both advantages of optical contrast and acoustic resolution, can visualize the chemical-specific optical information of tissues inside human-body. Recently, its corresponding reconstruction methods have been extensively researched. However, most of them are limited on cylindrical scan trajectories, rather than a helical scan which is more clinically practical. On this note, this article proposes a methodology of imaging reconstruction and evaluation for helical scan guided PAE. Different from traditional reconstruction method, synthetic aperture focusing technique (SAFT), our method reconstructs image using wavefield extrapolation which significantly improves computational efficiency and even takes only 0.25 seconds for 3-D reconstructions. In addition, the proposed evaluation methodology can estimate the resolutions and deviations of reconstructed images in advance, and then can be used to optimize the PAE scan parameters. Groups of simulations as well as ex-vivo experiments with different scan parameters are provided to fully demonstrate the performance of the proposed techniques. The quantitatively measured angular resolutions and deviations agree well with our theoretical derivation results ${D}{\sqrt {r_{s}^{2} + \overline {h}^{{2}}} } / {[{1.25}({r}_{s} {r}_{d} + \overline {h}^{{2}})] }$ (rad) and $- \overline {h} {l} / ({r}_{s} {r}_{d} + \overline {h}^{{2}})$ (rad), respectively ${D},{r}_{d},\;{r}_{s},\overline {h} $ and ${l}$ represent transducer diameter, radius of scan trajectory, radius of source position, unit helical pitch and the distance from targets to helical scan plane, respectively). This theoretical result also suits for circular and cylindrical scan in case of $\overline {h} = {0}$ .

A photoacoustic signal is proportional to a product of optical absorption coefficient and local light fluence; quantitative photoacoustic measurements of the optical absorption coefficients therefore require an accurate compensation of optical fluence variations. Usually, a supplementary diffuse optical tomography (DOT) is required to compensate the light fluence variations, but it is troubled by the bulky measurement system. In this paper, we propose a miniaturized dual-modality photoacoustic fusion imaging system. It utilizes an ultrasound sensor to receive photoacoustic signals for common photoacoustic imaging and extracts the passive ultrasound (PU) waves, backscattered laser-induced ultrasound waves, from the photoacoustic waves for diffuse reflectance (DR) imaging, simultaneously. Based on the dual-modality images of photoacoustic and DR, a further fusion photoacoustic imaging is implemented with the compensation the optical fluence variations. According to the experimental results, this miniaturized system can greatly reduce nearly 70% errors caused by the light fluence variations.

Diabetes mellitus is a prevalent metabolic disease which requires daily glucose monitoring for patients. Commercially available devices are all finger-prick which is painful and costly. Herein, we developed a portable and noninvasive glucose sensor based on photoacoustic (PA) technique. The size of the sensor main body is 17cm(L) × 14cm(W) × 10cm(H) with a weight of 650g (including a power bank). The system configuration is shown and the performance was evaluated by in vitro experiment on aqueous glucose solution at both high (1~8g/dL) and low concentrations (50~350 mg/dL). The R2and RMSE values achieved are 0.9526, 0.5903g/dL and 0.8966, 40.18mg/dL, respectively. Moreover, Clarke Error Grid analysis is also applied for physiological concentration prediction evaluation and 71.43% of data points fall in region A and the remaining points are in B, indicating the potential of our portable PA glucose sensor for clinical application.

Recently, many reconstruction methods have been developed to improve the lateral resolution of acoustic-resolution photoacoustic microscopy (ARPAM) in out-of-focus regions. Though these methods enhance image resolution to some extent, they require advanced computational hardware and large computational time, especially for three-dimensional (3-D) cases. However, some methods do not consider the finite size of a transducer, while others employ numerical discretization to build a focused transducer model that is less efficient and accurate. To overcome these problems, we propose a 3-D ARPAM imaging reconstruction method with high precision, high efficiency, and low memory cost. It inherits the framework of model-based reconstructions and incorporates the forward acoustic model in the hybrid domain. This hybrid-domain acoustic model promotes an analytical solution to establish a focused transducer model. Furthermore, the non-uniform fast Fourier transform (NUFFT) and deconvolution methods are introduced to reduce the required computational time and memory volume for 3-D reconstructions. According to the experimental results reconstructed by the proposed method, the lateral resolution of an ARPAM image recorded by a 20-MHz focused transducer (NA 0.393) can reach 88.39 [Formula: see text]. This resolution exceeds the diffraction limitation of the focused transducer ( [Formula: see text]). When reconstructing a 3-D image with 200×200×150 pixels, the proposed method takes only 8.15 s using a laptop loaded with Intel Core i7-8550U CPU at 1.8 GHz and 1.06-GB memory.

A handheld approach to 3D photoacoustic imaging is essential in clinical applications. To this end, we develop a 3D handheld photoacoustic imager for dynamic (temporally and spatially) volumetric visualization. In this 3D imager, the optically transmitting part and the acoustically receiving part are integrated into a single handheld probe with a compact size about 160 mm ×64 mm ×40 mm. Besides, a dedicated imaging reconstruction algorithm for the heterogeneous medium is developed based on the phase-shift migration method in the frequency domain, which deals well with the stratified condition in the designed system. Dynamic 3D imaging supporting flexible handheld operation is demonstrated with needle biopsy and in vitro temperature measurement for photothermal therapy. The development of such a 3D handheld photoacoustic system paves the way for compact and handheld-operating implementations, and its further clinical exploration is promising.

A novel noninvasive microwave biosensor for glucose measurement was designed, developed and tested to show its potential for real-time and continuous glucose monitoring. The reflected S-parameter S11 was investigated with different glucose concentrations within the range of 0.5 ~ 7 g/dL. The magnitude and resonance frequency information were utilized for prediction and then combined by data fusion to enhance the accuracy. The prediction results by magnitude and resonance frequency of S11 versus reference glucose concentration show root-mean-squared errors (RMSE) of 0.2887 g/dL and 0.1681 g/dL respectively. The prediction by data fusion shows RMSE value of 0.1364 g/dL indicating significant accuracy improvement of 52.75% and 18.86% compared to single parameter based magnitude and frequency predictions.