arXiv subjects

Yu Luo

Publications and source records attributed to Yu Luo.

At least 127 records · Page 7Linked to original sources

arXiv2024

Adaptive Mixed-Scale Feature Fusion Network for Blind AI-Generated Image Quality Assessment

With the increasing maturity of the text-to-image and image-to-image generative models, AI-generated images (AGIs) have shown great application potential in advertisement, entertainment, education, social media, etc. Although remarkable advancements have been achieved in generative models, very few efforts have been paid to design relevant quality assessment models. In this paper, we propose a novel blind image quality assessment (IQA) network, named AMFF-Net, for AGIs. AMFF-Net evaluates AGI quality from three dimensions, i.e., "visual quality", "authenticity", and "consistency". Specifically, inspired by the characteristics of the human visual system and motivated by the observation that "visual quality" and "authenticity" are characterized by both local and global aspects, AMFF-Net scales the image up and down and takes the scaled images and original-sized image as the inputs to obtain multi-scale features. After that, an Adaptive Feature Fusion (AFF) block is used to adaptively fuse the multi-scale features with learnable weights. In addition, considering the correlation between the image and prompt, AMFF-Net compares the semantic features from text encoder and image encoder to evaluate the text-to-image alignment. We carry out extensive experiments on three AGI quality assessment databases, and the experimental results show that our AMFF-Net obtains better performance than nine state-of-the-art blind IQA methods. The results of ablation experiments further demonstrate the effectiveness of the proposed multi-scale input strategy and AFF block.

cs.CV↗

arXiv2024

CSST Strong Lensing Preparation: a Framework for Detecting Strong Lenses in the Multi-color Imaging Survey by the China Survey Space Telescope (CSST)

Strong gravitational lensing is a powerful tool for investigating dark matter and dark energy properties. With the advent of large-scale sky surveys, we can discover strong lensing systems on an unprecedented scale, which requires efficient tools to extract them from billions of astronomical objects. The existing mainstream lens-finding tools are based on machine learning algorithms and applied to cut-out-centered galaxies. However, according to the design and survey strategy of optical surveys by CSST, preparing cutouts with multiple bands requires considerable efforts. To overcome these challenges, we have developed a framework based on a hierarchical visual Transformer with a sliding window technique to search for strong lensing systems within entire images. Moreover, given that multi-color images of strong lensing systems can provide insights into their physical characteristics, our framework is specifically crafted to identify strong lensing systems in images with any number of channels. As evaluated using CSST mock data based on an Semi-Analytic Model named CosmoDC2, our framework achieves precision and recall rates of 0.98 and 0.90, respectively. To evaluate the effectiveness of our method in real observations, we have applied it to a subset of images from the DESI Legacy Imaging Surveys and media images from Euclid Early Release Observations. 61 new strong lensing system candidates are discovered by our method. However, we also identified false positives arising primarily from the simplified galaxy morphology assumptions within the simulation. This underscores the practical limitations of our approach while simultaneously highlighting potential avenues for future improvements.

astro-ph.IM↗

arXiv2024

Simulating emission line galaxies for the next generation of large-scale structure surveys

We investigate emission line galaxies across cosmic time by combining the modified L-Galaxies semi-analytical galaxy formation model with the JiuTian cosmological simulation. We improve the tidal disruption model of satellite galaxies in L-Galaxies to address the time dependence problem. We utilise the public code CLOUDY to compute emission line ratios for a grid of HII region models. The emission line models assume the same initial mass function as that used to generate the spectral energy distribution of semi-analytical galaxies, ensuring a coherent treatment for modelling the full galaxy spectrum. By incorporating these emission line ratios with galaxy properties, we reproduce observed luminosity functions for H$α$, H$β$, [OII], and [OIII] in the local Universe and at high redshifts. We also find good agreement between model predictions and observations for auto-correlation and cross-correlation functions of [OII]-selected galaxies, as well as their luminosity dependence. The bias of emission line galaxies depends on both luminosity and redshift. At lower redshifts, it remains constant with increasing luminosity up to around $\sim 10^{42.5}\rm \, erg\,s^{-1}$ and then rises steeply for higher luminosities. The transition luminosity increases with redshift and becomes insignificant above $z$=1.5. Generally, galaxy bias shows an increasing trend with redshift. However, for luminous galaxies, the bias is higher at low redshifts, as the strong luminosity dependence observed at low redshifts diminishes at higher redshifts. We provide a fitting formula for the bias of emission line galaxies as a function of luminosity and redshift, which can be utilised for large-scale structure studies with future galaxy surveys.

astro-ph.GA↗

arXiv2024

Forklens: Accurate weak-lensing shear measurement with deep learning

Weak gravitational lensing is one of the most important probes of the nature of dark matter and dark energy. In order to extract cosmological information from next-generation weak lensing surveys (e.g., Euclid, Roman, LSST, and CSST) as much as possible, accurate measurements of weak lensing shear are required. There are existing algorithms to measure the weak lensing shear on imaging data, which have been successfully applied in previous surveys. In the meantime, machine learning (ML) has been widely recognized in various astrophysics applications in modeling and observations. In this work, we present a fully deep-learning-based approach to measuring weak lensing shear accurately. Our approach comprises two modules. The first one contains a convolutional neural network (CNN) with two branches for taking galaxy images and point spread function (PSF) simultaneously, and the output of this module includes the galaxy's magnitude, size, and shape. The second module includes a multiple-layer neural network (NN) to calibrate weak-lensing shear measurements. We name the program Forklens and make it publicly available online. Applying Forklens to CSST-like mock images, we achieve consistent accuracy with traditional approaches (such as moment-based measurement and forward model fitting) on the sources with high signal-to-noise ratios (S/N, > 20). For the sources with S/N < 10, Forklens exhibits an $\sim 36\%$ higher Pearson coefficient on galaxy ellipticity measurements. After adopting galaxy weighting, the shear measurements with Forklens deliver accuracy levels to $0.2\%$. The whole procedure of Forklens is automated and costs about $0.7$ milliseconds per galaxy, which is appropriate for adequately taking advantage of the sky coverage and depth of the upcoming weak lensing surveys.

astro-ph.CO↗

arXiv2024

CSST large-scale structure analysis pipeline: I. constructing reference mock galaxy redshift surveys

In this paper, we set out to construct a set of reference mock galaxy redshift surveys (MGRSs) for the future Chinese Space-station Survey Telescope (CSST) observation, where subsequent survey selection effects can be added and evaluated. This set of MGRSs is generated using the dark matter subhalos extracted from a high-resolution Jiutian $N$-body simulation of the standard $Λ$CDM cosmogony with $Ω_m=0.3111$, $Ω_Λ=0.6889$, and $σ_8=0.8102$. The simulation has a boxsize of $1~h^{-1} {\rm Gpc}$, and consists of $6144^3$ particles with mass resolution $3.723 \times 10^{8} h^{-1} M_\odot $. In order to take into account the effect of redshift evolution, we first use all 128 snapshots in the Jiutian simulation to generate a light-cone halo/subhalo catalog. Next, galaxy luminosities are assigned to the main and subhalo populations using the subhalo abundance matching (SHAM) method with the DESI $z$-band luminosity functions at different redshifts. Multi-band photometries, as well as images, are then assigned to each mock galaxy using a 3-dimensional parameter space nearest neighbor sampling of the DESI LS observational galaxies and groups. Finally, the CSST and DESI LS survey geometry and magnitude limit cuts are applied to generate the required MGRSs. As we have checked, this set of MGRSs can generally reproduce the observed galaxy luminosity/mass functions within 0.1 dex for galaxies with $L > 10^8 L_\odot$ (or $M_* > 10^{8.5} M_\odot$) and within 1-$σ$ level for galaxies with $L < 10^8L_\odot$ (or $M_* < 10^{8.5} M_\odot$). Together with the CSST slitless spectra and redshifts for our DESI LS seed galaxies that are under construction, we will set out to test various slitless observational selection effects in subsequent probes.

astro-ph.GA↗

arXiv2024

Catalogue of nearby blue and near-solar gas metallicity SDSS dwarf galaxies

A less explored aspect of dwarf galaxies is their metallicity evolution. Generally, dwarfs have lower metallicities than Hubble sequence late type galaxies but in reality, dwarfs span a wide range of metallicities with several open questions regarding the formation and evolution of the lowest and the highest metallicity dwarfs. We present a catalogue of 3459 blue, nearby, star forming dwarf galaxies extracted from SDSS DR16 including calculation of their metallicities using the mean of several calibrators. To compile our catalogue we applied redshift, absolute magnitude, stellar mass, optical diameter, and line flux signal to noise criteria. This produced a catalogue from the upper end of the dwarf galaxy stellar mass range. Our catalogued dwarfs have blue g - i colours and Hbeta equivalent widths, indicative of having undergone a recent episode of star formation, although their star formation rates (SFR) suggest only a moderate to low enhancement in star formation, similar to the SFRs in low surface brightness and evolved tidal dwarfs. While the catalogued dwarfs cover a range of metallicities, their mean metallicity is about 0.2 dex below solar metallicity, indicating relatively chemically evolved galaxies. The vast majority of the catalogue, with clean photometry, are relatively isolated dwarfs with only modest star formation rates and a narrow range of g - i colour, consistent with internally driven episodic mild bursts of star formation. The presented catalogue's robust metallicity estimates for nearby SDSS dwarf galaxies will help target future studies to understand the physical processes driving the metallicity evolution of dwarfs.

astro-ph.GA↗

arXiv2024

DrM: Mastering Visual Reinforcement Learning through Dormant Ratio Minimization

Visual reinforcement learning (RL) has shown promise in continuous control tasks. Despite its progress, current algorithms are still unsatisfactory in virtually every aspect of the performance such as sample efficiency, asymptotic performance, and their robustness to the choice of random seeds. In this paper, we identify a major shortcoming in existing visual RL methods that is the agents often exhibit sustained inactivity during early training, thereby limiting their ability to explore effectively. Expanding upon this crucial observation, we additionally unveil a significant correlation between the agents' inclination towards motorically inactive exploration and the absence of neuronal activity within their policy networks. To quantify this inactivity, we adopt dormant ratio as a metric to measure inactivity in the RL agent's network. Empirically, we also recognize that the dormant ratio can act as a standalone indicator of an agent's activity level, regardless of the received reward signals. Leveraging the aforementioned insights, we introduce DrM, a method that uses three core mechanisms to guide agents' exploration-exploitation trade-offs by actively minimizing the dormant ratio. Experiments demonstrate that DrM achieves significant improvements in sample efficiency and asymptotic performance with no broken seeds (76 seeds in total) across three continuous control benchmark environments, including DeepMind Control Suite, MetaWorld, and Adroit. Most importantly, DrM is the first model-free algorithm that consistently solves tasks in both the Dog and Manipulator domains from the DeepMind Control Suite as well as three dexterous hand manipulation tasks without demonstrations in Adroit, all based on pixel observations.

cs.LG↗

arXiv2024

A conformal mapping approach to broadband nonlinear optics on chip

Integrated nonlinear optical devices play an important role in modern optical communications. However, conventional on-chip optical devices with homogeneous or periodic translation dimensions generally have limited bandwidth when applied to nonlinear optical applications. Up today, there lacks a general method to design compact nonlinear optical devices over a broadband continuous frequency range. In this work, we propose a general strategy based on transformation optics (TO) to design curved accelerating waveguides (CAWs) with spatially gradient curvatures able to achieve broadband nonlinear frequency conversion on chip. Through rigorous analytical calculation, we show that increasing the acceleration (i.e. gradient in the waveguide curvature) broadens the output signal spectrum in the nonlinear process. In the experiment, we take the sum-frequency generation for infrared signal upconversion (SFG-ISU) as the example and fabricated a variety of CAWs using thin-film lithium niobate on insulator (LNOI). Efficient SFG is observed over a broadband continuous spectrum. Our conformal mapping approach offers a platform for various nonlinear optical processes and works in any frequency range, including visible, infrared and terahertz bands. Apart from LNOI, our approach is also compatible with other nonlinear materials, such as silicon, silicon nitride and chalcogenide glasses etc.

physics.optics↗

arXiv2024

Recurrence relations of Exceptional Laurent biorthogonal polynomials

Exceptional extensions of a class of Laurent biorthogonal polynomials (the so-called Hendriksen-van Rossum polynomials) have been presented by the authors recently. This is achieved through Darboux transformations of generalized eigenvalue problems. In this paper, we discuss the recurrence relations satisfied by these exceptional Laurent biorthogonal polynomials and provide a type of recurrence relations with $3l_0+4$ terms explicitly, where the parameter $l_0$ corresponds to the degree of the polynomial part in the seed function used in the Darboux transformation. In the proof of these recurrence relations, the backward operator which maps an exceptional polynomial into a classical one plays a significant role.

math.CA↗

arXiv2024

CBVS: A Large-Scale Chinese Image-Text Benchmark for Real-World Short Video Search Scenarios

Vision-Language Models pre-trained on large-scale image-text datasets have shown superior performance in downstream tasks such as image retrieval. Most of the images for pre-training are presented in the form of open domain common-sense visual elements. Differently, video covers in short video search scenarios are presented as user-originated contents that provide important visual summaries of videos. In addition, a portion of the video covers come with manually designed cover texts that provide semantic complements. In order to fill in the gaps in short video cover data, we establish the first large-scale cover-text benchmark for Chinese short video search scenarios. Specifically, we release two large-scale datasets CBVS-5M/10M to provide short video covers, and the manual fine-labeling dataset CBVS-20K to provide real user queries, which serves as an image-text benchmark test in the Chinese short video search field. To integrate the semantics of cover text in the case of modality missing, we propose UniCLIP where cover texts play a guiding role during training, however are not relied upon by inference. Extensive evaluation on CBVS-20K demonstrates the excellent performance of our proposal. UniCLIP has been deployed to Tencent's online video search systems with hundreds of millions of visits and achieved significant gains. The dataset and code are available at https://github.com/QQBrowserVideoSearch/CBVS-UniCLIP.

cs.CV↗

arXiv2023

Quantitative perfusion maps using a novelty spatiotemporal convolutional neural network

Dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) is widely used to evaluate acute ischemic stroke to distinguish salvageable tissue and infarct core. For this purpose, traditional methods employ deconvolution techniques, like singular value decomposition, which are known to be vulnerable to noise, potentially distorting the derived perfusion parameters. However, deep learning technology could leverage it, which can accurately estimate clinical perfusion parameters compared to traditional clinical approaches. Therefore, this study presents a perfusion parameters estimation network that considers spatial and temporal information, the Spatiotemporal Network (ST-Net), for the first time. The proposed network comprises a designed physical loss function to enhance model performance further. The results indicate that the network can accurately estimate perfusion parameters, including cerebral blood volume (CBV), cerebral blood flow (CBF), and time to maximum of the residual function (Tmax). The structural similarity index (SSIM) mean values for CBV, CBF, and Tmax parameters were 0.952, 0.943, and 0.863, respectively. The DICE score for the hypo-perfused region reached 0.859, demonstrating high consistency. The proposed model also maintains time efficiency, closely approaching the performance of commercial gold-standard software.

eess.IV↗

arXiv2023

Efficient LLM Inference on CPUs

Large language models (LLMs) have demonstrated remarkable performance and tremendous potential across a wide range of tasks. However, deploying these models has been challenging due to the astronomical amount of model parameters, which requires a demand for large memory capacity and high memory bandwidth. In this paper, we propose an effective approach that can make the deployment of LLMs more efficiently. We support an automatic INT4 weight-only quantization flow and design a special LLM runtime with highly-optimized kernels to accelerate the LLM inference on CPUs. We demonstrate the general applicability of our approach on popular LLMs including Llama2, Llama, GPT-NeoX, and showcase the extreme inference efficiency on CPUs. The code is publicly available at: https://github.com/intel/intel-extension-for-transformers.

cs.LG↗

arXiv2023

When to Update Your Model: Constrained Model-based Reinforcement Learning

Designing and analyzing model-based RL (MBRL) algorithms with guaranteed monotonic improvement has been challenging, mainly due to the interdependence between policy optimization and model learning. Existing discrepancy bounds generally ignore the impacts of model shifts, and their corresponding algorithms are prone to degrade performance by drastic model updating. In this work, we first propose a novel and general theoretical scheme for a non-decreasing performance guarantee of MBRL. Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. These discoveries encourage us to formulate a constrained lower-bound optimization problem to permit the monotonicity of MBRL. A further example demonstrates that learning models from a dynamically-varying number of explorations benefit the eventual returns. Motivated by these analyses, we design a simple but effective algorithm CMLO (Constrained Model-shift Lower-bound Optimization), by introducing an event-triggered mechanism that flexibly determines when to update the model. Experiments show that CMLO surpasses other state-of-the-art methods and produces a boost when various policy optimization methods are employed.

cs.LG↗

arXiv2023

Accelerating Training of MLIPs Through Small-Cell Training

While machine-learned interatomic potentials have become a mainstay for modeling materials, designing training sets that lead to robust potentials is challenging. Automated methods, such as active learning and on-the-fly learning, construct reliable training sets, but these processes can be resource-intensive. Current training approaches often use density functional theory (DFT) calculations that have the same cell size as the simulations that the potential is explicitly trained to model. Here, we demonstrate an easy-to-implement small-cell training protocol and use it to model the Zr-H system. This training leads to a potential that accurately predicts known stable Zr-H phases and reproduces the $α$-$β$ pure zirconium phase transition in molecular dynamics simulations. Compared to traditional active learning, small-cell training decreased the training time of the $α$-$β$ zirconium phase transition by approximately 20 times. The potential describes the phase transition with a degree of accuracy similar to that of the large-cell training method.

cond-mat.mtrl-sci↗

arXiv2023

Anisotropic diffusion of radiation-induced self-interstitial clusters in HCP zirconium: a molecular dynamics and rate-theory assessment

Under irradiation, Zr and Zr alloys undergo growth in the absence of applied stress. This phenomenon is thought to be associated with the anisotropy of diffusion of either or both radiation-induced point defects and defect clusters. In this work, molecular dynamic simulations are used to study the anisotropy of diffusion of self-interstitial atom clusters. Both near-equilibrium clusters generated by aggregation of self-interstitial atoms and cascade-induced clusters were considered. The cascade-induced clusters display more anisotropy than their counterparts produced by aggregation. In addition to 1-dimensional diffusing clusters, 2-dimensional diffusing clusters were observed. Using our molecular dynamic simulations, the input parameters for the "self-interstitial atom cluster bias" rate-theory model were estimated. The radiation-induced growth strains predicted using this model are largely consistent with experiments, but are highly sensitive to the choice of interatomic interaction potential.

cond-mat.mtrl-sci↗

arXiv2023

Simulation Of The Microlensing Effect Near The Critical Curve Of The Galaxy Cluster

In the smooth mass distribution model, the critical curve represents a line with magnification divergence on the image plane in a strong gravitational lensing system. Considering the microlensing effects caused by discrete masses, the magnification map in the source plane exhibits a complex structure, which offers a promising way for detecting dark matter. However, simulating microlensing near the critical curve poses challenges due to magnification divergence and the substantial computational demands involved. To achieve the required simulation accuracy, direct inverse ray-shooting would require significant computational resources. Therefore we applied a GPU-based code optimized with interpolation method to enable efficient computation on a large scale. Using the GPU of NVIDIA Tesla V100S PCIe 32GB, it takes approximately 7000 seconds to calculate the effects of around 13,000 microlenses for a simulation involving 1013 emitted rays. Then we generated 80 magnification maps, and select 800 light curves for a statistical analysis of microcaustic density and peak magnification.

astro-ph.GA↗

arXiv2023

Enhancing In-Situ Structural Health Monitoring through RF Energy-Powered Sensor Nodes and Mobile Platform

This research contributes to long-term structural health monitoring (SHM) by exploring radio frequency energy-powered sensor nodes (RF-SNs) embedded in concrete. Unlike traditional in-situ monitoring systems relying on batteries or wire-connected power sources, the RF-SN captures radio energy from a mobile radio transmitter for sensing and communication. This offers a cost-effective solution for consistent in-situ perception. To optimize the system performance across various situations, we've explored both active and passive communication methods. For the active RF-SN, we implement a specialized control circuit enabling the node to transmit data through ZigBee protocol at low incident power. For the passive RF-SN, radio energy is not only for power but also as a carrier signal, with data conveyed by modulating the amplitude of the backscattered radio wave. To address the challenge of significant attenuation of the backscattering signal in concrete, we utilize a square chirp-based modulation scheme for passive communication. This scheme allows the receiver to successfully decode the data even under a negative signal-to-noise ratio (SNR) condition. The experimental results indicate that an active RF-SN embedded in concrete at a depth of 13.5 cm can be effectively powered by a 915MHz mobile radio transmitter with an effective isotropic radiated power (EIRP) of 32.5dBm. This setup allows the RF-SN to send over 1 kilobyte of data within 10 seconds, with an additional 1.7 kilobytes every 1.6 seconds of extra charging. For the passive RF-SN buried at the same depth, continuous data transmission at a rate of 224 bps with a 3% bit error rate (BER) is achieved when the EIRP of the transmitter is 23.6 dBm.

eess.SY↗

arXiv2023

An Efficient Sparse Inference Software Accelerator for Transformer-based Language Models on CPUs

In recent years, Transformer-based language models have become the standard approach for natural language processing tasks. However, stringent throughput and latency requirements in industrial applications are limiting their adoption. To mitigate the gap, model compression techniques such as structured pruning are being used to improve inference efficiency. However, most existing neural network inference runtimes lack adequate support for structured sparsity. In this paper, we propose an efficient sparse deep learning inference software stack for Transformer-based language models where the weights are pruned with constant block size. Our sparse software accelerator leverages Intel Deep Learning Boost to maximize the performance of sparse matrix - dense matrix multiplication (commonly abbreviated as SpMM) on CPUs. Our SpMM kernel outperforms the existing sparse libraries (oneMKL, TVM, and LIBXSMM) by an order of magnitude on a wide range of GEMM shapes under 5 representative sparsity ratios (70%, 75%, 80%, 85%, 90%). Moreover, our SpMM kernel shows up to 5x speedup over dense GEMM kernel of oneDNN, a well-optimized dense library widely used in industry. We apply our sparse accelerator on widely-used Transformer-based language models including Bert-Mini, DistilBERT, Bert-Base, and BERT-Large. Our sparse inference software shows up to 1.5x speedup over Neural Magic's Deepsparse under same configurations on Xeon on Amazon Web Services under proxy production latency constraints. We also compare our solution with two framework-based inference solutions, ONNX Runtime and PyTorch, and demonstrate up to 37x speedup over ONNX Runtime and 345x over PyTorch on Xeon under the latency constraints. All the source code is publicly available on Github: https://github.com/intel/intel-extension-for-transformers.

cs.LG↗