arXiv subjects

Yuhang Li

Publications and source records attributed to Yuhang Li.

At least 73 records · Page 4Linked to original sources

arXiv2025

Snapshot multi-spectral imaging through defocusing and a Fourier imager network

Multi-spectral imaging, which simultaneously captures the spatial and spectral information of a scene, is widely used across diverse fields, including remote sensing, biomedical imaging, and agricultural monitoring. Here, we introduce a snapshot multi-spectral imaging approach employing a standard monochrome image sensor with no additional spectral filters or customized components. Our system leverages the inherent chromatic aberration of wavelength-dependent defocusing as a natural source of physical encoding of multi-spectral information; this encoded image information is rapidly decoded via a deep learning-based multi-spectral Fourier Imager Network (mFIN). We experimentally tested our method with six illumination bands and demonstrated an overall accuracy of 92.98% for predicting the illumination channels at the input and achieved a robust multi-spectral image reconstruction on various test objects. This deep learning-powered framework achieves high-quality multi-spectral image reconstruction using snapshot image acquisition with a monochrome image sensor and could be useful for applications in biomedicine, industrial quality control, and agriculture, among others.

physics.optics↗

arXiv2025

Roadmap on Neuromorphic Photonics

This roadmap consolidates recent advances while exploring emerging applications, reflecting the remarkable diversity of hardware platforms, neuromorphic concepts, and implementation philosophies reported in the field. It emphasizes the critical role of cross-disciplinary collaboration in this rapidly evolving field.

cs.ET↗

arXiv2024

Maximum Principle for Control System driven by Mixed Fractional Brownian Motion

In this paper, we investigate the optimal control problem for systems driven by mixed fractional Brownian motion (including a fractional Brownian motion with Hurst parameter $H>1/2$ and the standard Brownian motion). By using Malliavin calculus and introducing a disturbance control region, we obtain a modified maximum principle. Through martingale representation theorem, we obtain the adjoint backward stochastic differential equation in a natural way. Furthermore, corresponding to [1], a significant result is that the necessary condition is simplified by only containing one equality. As an application, the linear quadratic case is investigated to illustrate the main results.

math.OC↗

arXiv2024

Maximum principle for discrete-time control systems driven by fractional noises and related backward stochastic difference equations

In this paper, the optimal control for discrete-time systems driven by fractional noises is studied. A stochastic maximum principle is obtained by introducing a backward stochastic difference equation contains both fractional noises and the constructed white noises. The solution of the backward stochastic difference equations is also investigated. As an application, the linear quadratic case is considered to illustrate the main results.

math.OC↗

arXiv2024

Linear Filtering for Discrete Time Systems Driven by Fractional Noises

In this paper, we study the discrete time filtering problems for linear systems driven by fractional noises. The main difficulty comes from the non-Markovian of the noises. We construct the difference equation of the covariance process through the properties of the noises and transform the filtering problem to an optimal control problem. We obtain the necessary condition that the coefficients of the optimal filter should satisfy and show there is no consistent optimal filter, which is a significant difference from the classical Kalman filter. Finally, a simple example is considered to illustrate the main results. Further more, our method could also deal with systems driven by any other colored noises, as long as the self-covariation function is known.

math.OC↗

arXiv2024

ITPNet: Towards Instantaneous Trajectory Prediction for Autonomous Driving

Trajectory prediction of agents is crucial for the safety of autonomous vehicles, whereas previous approaches usually rely on sufficiently long-observed trajectory to predict the future trajectory of the agents. However, in real-world scenarios, it is not realistic to collect adequate observed locations for moving agents, leading to the collapse of most prediction models. For instance, when a moving car suddenly appears and is very close to an autonomous vehicle because of the obstruction, it is quite necessary for the autonomous vehicle to quickly and accurately predict the future trajectories of the car with limited observed trajectory locations. In light of this, we focus on investigating the task of instantaneous trajectory prediction, i.e., two observed locations are available during inference. To this end, we propose a general and plug-and-play instantaneous trajectory prediction approach, called ITPNet. Specifically, we propose a backward forecasting mechanism to reversely predict the latent feature representations of unobserved historical trajectories of the agent based on its two observed locations and then leverage them as complementary information for future trajectory prediction. Meanwhile, due to the inevitable existence of noise and redundancy in the predicted latent feature representations, we further devise a Noise Redundancy Reduction Former, aiming at to filter out noise and redundancy from unobserved trajectories and integrate the filtered features and observed features into a compact query for future trajectory predictions. In essence, ITPNet can be naturally compatible with existing trajectory prediction models, enabling them to gracefully handle the case of instantaneous trajectory prediction. Extensive experiments on the Argoverse and nuScenes datasets demonstrate ITPNet outperforms the baselines, and its efficacy with different trajectory prediction models.

cs.CV↗

arXiv2024

Unidirectional focusing of light using structured diffractive surfaces

Unidirectional optical systems enable selective control of light through asymmetric processing of radiation, effectively transmitting light in one direction while blocking unwanted propagation in the opposite direction. Here, we introduce a reciprocal diffractive unidirectional focusing design based on linear and isotropic diffractive layers that are structured. Using deep learning-based optimization, a cascaded set of diffractive layers are spatially engineered at the wavelength scale to focus light efficiently in the forward direction while blocking it in the opposite direction. The forward energy focusing efficiency and the backward energy suppression capabilities of this unidirectional architecture were demonstrated under various illumination angles and wavelengths, illustrating the versatility of our polarization-insensitive design. Furthermore, we demonstrated that these designs are resilient to adversarial attacks that utilize wavefront engineering from outside. Experimental validation using terahertz radiation confirmed the feasibility of this diffractive unidirectional focusing framework. Diffractive unidirectional designs can operate across different parts of the electromagnetic spectrum by scaling the resulting diffractive features proportional to the wavelength of light and will find applications in security, defense, and optical communication, among others.

physics.optics↗

arXiv2024

Machine learning enhanced multi-particle tracking in solid fuel combustion

Particle velocimetry is essential in solid fuel combustion studies, however, the accurate detection and tracking of particles in high Particle Number Density (PND) combustion scenario remain challenging. The current study advances the machine-learning approaches for precise velocity measurements of solid particles. For this, laser imaging experiments were performed for high-volatile bituminous coal particles burning in a laminar flow reactor. Particle positions were imaged using time-resolved Mie scattering. Various detection methods, including conventional blob detection and Machine Learning (ML) based You Only Look Once (YOLO) and Realtime Detection Transformer (RT-DETR) were employed and bench marked.~Particle tracking was performed using the Simple Online Realtime Tracking (SORT) algorithm. The results demonstrated the capability of machine learning models trained on low-PND data for prediction of high-PND data. Slicing Aided Hyper Inference (SAHI) algorithm is important for the better performance of the used models. By evaluating the velocity statistics, it is found that the mean particle velocity decreases with increasing PND, primarily due to stronger particle interactions. The particle dynamics are closely related to the position of combustion zone observed in the previous study. Thus, PND is considered as the dominant factor for the particle group combustion behavior of high-volatile solid fuels.

physics.app-ph↗

arXiv2024

Large non-Gaussianities corresponding to first-order phase transitions during inflation

In this study, we explore the back reaction of phase transitions in the spectator sector on the inflaton field during slow-roll inflation. Due to the significant excursion of the inflaton field, these phase transitions are likely to occur and can induce substantial non-Gaussian correlations in the curvature perturbation. Our results suggest that these correlations could be detectable by future observations of the cosmic microwave background radiation and large-scale structure surveys. Furthermore, we demonstrate that in certain parameter spaces, a scaling non-Gaussian signal can be produced, offering deeper insights into both the inflaton and spectator sectors. Additionally, phase transitions during inflation can generate gravitational wave signals with distinctive signatures, potentially explaining observations made by pulsar timing array experiments. The associated non-Gaussian correlations provide collateral evidence for these phase transitions.

hep-ph↗

arXiv2024

PCoTTA: Continual Test-Time Adaptation for Multi-Task Point Cloud Understanding

In this paper, we present PCoTTA, an innovative, pioneering framework for Continual Test-Time Adaptation (CoTTA) in multi-task point cloud understanding, enhancing the model's transferability towards the continually changing target domain. We introduce a multi-task setting for PCoTTA, which is practical and realistic, handling multiple tasks within one unified model during the continual adaptation. Our PCoTTA involves three key components: automatic prototype mixture (APM), Gaussian Splatted feature shifting (GSFS), and contrastive prototype repulsion (CPR). Firstly, APM is designed to automatically mix the source prototypes with the learnable prototypes with a similarity balancing factor, avoiding catastrophic forgetting. Then, GSFS dynamically shifts the testing sample toward the source domain, mitigating error accumulation in an online manner. In addition, CPR is proposed to pull the nearest learnable prototype close to the testing feature and push it away from other prototypes, making each prototype distinguishable during the adaptation. Experimental comparisons lead to a new benchmark, demonstrating PCoTTA's superiority in boosting the model's transferability towards the continually changing target domain.

cs.CV↗

arXiv2024

TesseraQ: Ultra Low-Bit LLM Post-Training Quantization with Block Reconstruction

Large language models (LLMs) have revolutionized natural language processing, albeit at the cost of immense memory and computation requirements. Post-training quantization (PTQ) is becoming the de facto method to reduce the memory footprint and improve the inference throughput of LLMs. In this work, we aim to push the upper limit of LLM PTQ by optimizing the weight rounding parameters with the block reconstruction technique, a predominant method in previous vision models. We propose TesseraQ, a new state-of-the-art PTQ technique, to quantize the weights of LLMs to ultra-low bits. To effectively optimize the rounding in LLMs and stabilize the reconstruction process, we introduce progressive adaptive rounding. This approach iteratively transits the soft rounding variables to hard variables during the reconstruction process. Additionally, we optimize the dequantization scale parameters to fully leverage the block reconstruction technique. We demonstrate that TesseraQ can be seamlessly integrated with existing scaling or clipping-based PTQ algorithms such as AWQ and OmniQuant, significantly enhancing their performance and establishing a new state-of-the-art. For instance, when compared to AWQ, TesseraQ improves the wikitext2 perplexity from 14.65 to 6.82 and average downstream accuracy from 50.52 to 59.27 with 2-bit weight-only quantization of LLaMA-2-7B. Across a range of quantization schemes, including W2A16, W3A16, W3A3, and W4A4, TesseraQ consistently exhibits superior performance.

cs.LG↗

arXiv2024

Optical Generative Models

Generative models cover various application areas, including image, video and music synthesis, natural language processing, and molecular design, among many others. As digital generative models become larger, scalable inference in a fast and energy-efficient manner becomes a challenge. Here, we present optical generative models inspired by diffusion models, where a shallow and fast digital encoder first maps random noise into phase patterns that serve as optical generative seeds for a desired data distribution; a jointly-trained free-space-based reconfigurable decoder all-optically processes these generative seeds to create novel images (never seen before) following the target data distribution. Except for the illumination power and the random seed generation through a shallow encoder, these optical generative models do not consume computing power during the synthesis of novel images. We report the optical generation of monochrome and multi-color novel images of handwritten digits, fashion products, butterflies, and human faces, following the data distributions of MNIST, Fashion MNIST, Butterflies-100, and Celeb-A datasets, respectively, achieving an overall performance comparable to digital neural network-based generative models. To experimentally demonstrate optical generative models, we used visible light to generate, in a snapshot, novel images of handwritten digits and fashion products. These optical generative models might pave the way for energy-efficient, scalable and rapid inference tasks, further exploiting the potentials of optics and photonics for artificial intelligence-generated content.

cs.NE↗

arXiv2024

GenQ: Quantization in Low Data Regimes with Generative Synthetic Data

In the realm of deep neural network deployment, low-bit quantization presents a promising avenue for enhancing computational efficiency. However, it often hinges on the availability of training data to mitigate quantization errors, a significant challenge when data availability is scarce or restricted due to privacy or copyright concerns. Addressing this, we introduce GenQ, a novel approach employing an advanced Generative AI model to generate photorealistic, high-resolution synthetic data, overcoming the limitations of traditional methods that struggle to accurately mimic complex objects in extensive datasets like ImageNet. Our methodology is underscored by two robust filtering mechanisms designed to ensure the synthetic data closely aligns with the intrinsic characteristics of the actual training data. In case of limited data availability, the actual data is used to guide the synthetic data generation process, enhancing fidelity through the inversion of learnable token embeddings. Through rigorous experimentation, GenQ establishes new benchmarks in data-free and data-scarce quantization, significantly outperforming existing methods in accuracy and efficiency, thereby setting a new standard for quantization in low data regimes. Code is released at \url{https://github.com/Intelligent-Computing-Lab-Yale/GenQ}.

cs.CV↗

arXiv2024

Graph Neural Networks for Wireless Networks: Graph Representation, Architecture and Evaluation

Graph neural networks (GNNs) have been regarded as the basic model to facilitate deep learning (DL) to revolutionize resource allocation in wireless networks. GNN-based models are shown to be able to learn the structural information about graphs representing the wireless networks to adapt to the time-varying channel state information and dynamics of network topology. This article aims to provide a comprehensive overview of applying GNNs to optimize wireless networks via answering three fundamental questions, i.e., how to input the wireless network data into GNNs, how to improve the performance of GNNs, and how to evaluate GNNs. Particularly, two graph representations are given to transform wireless network parameters into graph-structured data. Then, we focus on the architecture design of the GNN-based models via introducing the basic message passing as well as model improvement methods including multi-head attention mechanism and residual structure. At last, we give task-oriented evaluation metrics for DL-enabled wireless resource allocation. We also highlight certain challenges and potential research directions for the application of GNNs in wireless networks.

eess.SP↗

arXiv2024

ReSpike: Residual Frames-based Hybrid Spiking Neural Networks for Efficient Action Recognition

Spiking Neural Networks (SNNs) have emerged as a compelling, energy-efficient alternative to traditional Artificial Neural Networks (ANNs) for static image tasks such as image classification and segmentation. However, in the more complex video classification domain, SNN-based methods fall considerably short of ANN-based benchmarks due to the challenges in processing dense frame sequences. To bridge this gap, we propose ReSpike, a hybrid framework that synergizes the strengths of ANNs and SNNs to tackle action recognition tasks with high accuracy and low energy cost. By decomposing film clips into spatial and temporal components, i.e., RGB image Key Frames and event-like Residual Frames, ReSpike leverages ANN for learning spatial information and SNN for learning temporal information. In addition, we propose a multi-scale cross-attention mechanism for effective feature fusion. Compared to state-of-the-art SNN baselines, our ReSpike hybrid architecture demonstrates significant performance improvements (e.g., >30% absolute accuracy improvement on HMDB-51, UCF-101, and Kinetics-400). Furthermore, ReSpike achieves comparable performance with prior ANN approaches while bringing better accuracy-energy tradeoff.

cs.CV↗

arXiv2024

Optimal Filtering for Interacting Particle Systems

In this paper, we study the optimal filtering problem for a interacting particle system generated by stochastic differential equations with interaction. By using Malliavin calculus, we construct the differential equation of the covariance process and transform the filter problem to an optimal control problem. Finally we give the necessary condition that the coefficient of the optimal filter should satisfy

math.OC↗

arXiv2024

When In-memory Computing Meets Spiking Neural Networks -- A Perspective on Device-Circuit-System-and-Algorithm Co-design

This review explores the intersection of bio-plausible artificial intelligence in the form of Spiking Neural Networks (SNNs) with the analog In-Memory Computing (IMC) domain, highlighting their collective potential for low-power edge computing environments. Through detailed investigation at the device, circuit, and system levels, we highlight the pivotal synergies between SNNs and IMC architectures. Additionally, we emphasize the critical need for comprehensive system-level analyses, considering the inter-dependencies between algorithms, devices, circuit & system parameters, crucial for optimal performance. An in-depth analysis leads to identification of key system-level bottlenecks arising from device limitations which can be addressed using SNN-specific algorithm-hardware co-design techniques. This review underscores the imperative for holistic device to system design space co-exploration, highlighting the critical aspects of hardware and algorithm research endeavors for low-power neuromorphic solutions.

cs.NE↗

arXiv2024

Unidirectional imaging with partially coherent light

Unidirectional imagers form images of input objects only in one direction, e.g., from field-of-view (FOV) A to FOV B, while blocking the image formation in the reverse direction, from FOV B to FOV A. Here, we report unidirectional imaging under spatially partially coherent light and demonstrate high-quality imaging only in the forward direction (A->B) with high power efficiency while distorting the image formation in the backward direction (B->A) along with low power efficiency. Our reciprocal design features a set of spatially engineered linear diffractive layers that are statistically optimized for partially coherent illumination with a given phase correlation length. Our analyses reveal that when illuminated by a partially coherent beam with a correlation length of ~1.5 w or larger, where w is the wavelength of light, diffractive unidirectional imagers achieve robust performance, exhibiting asymmetric imaging performance between the forward and backward directions - as desired. A partially coherent unidirectional imager designed with a smaller correlation length of less than 1.5 w still supports unidirectional image transmission, but with a reduced figure of merit. These partially coherent diffractive unidirectional imagers are compact (axially spanning less than 75 w), polarization-independent, and compatible with various types of illumination sources, making them well-suited for applications in asymmetric visual information processing and communication.

physics.optics↗