arXiv subjects

Yanqi Huang

Publications and source records attributed to Yanqi Huang.

11 recordsLinked to original sources

arXiv2026

Estimation and Inference for Latent Markov Models by Fourier Recursions

This paper proposes a new theoretically exact Fourier recursion framework for a broad class of latent Markov models (LMMs), encompassing models widely used across a broad range of fields in economics. It can be viewed as a counterpart of the celebrated Kalman filter for non-Gaussian and nonlinear LMMs. Closed-form recursive updates of Fourier coefficients jointly deliver filtering, likelihood evaluation, and simultaneously accumulate the score and Hessian online. We introduce a unified truncated implementation that ensures uniform error control and numerical stability, preventing approximation errors from accumulating through the recursion. We establish asymptotic properties of the feasible maximum likelihood estimator for LMMs, provide a recursion-based consistent estimator for the Fisher information and discuss the possible model misspecification. These results provide the first general asymptotic theory for feasible approximate maximum likelihood estimation in LMMs. Simulations demonstrate its accuracy and stability, while an application to U.S. bankruptcy data recovers a persistent latent bankruptcy-pressure process.

stat.ME↗

arXiv2025

Likelihood Matching for Diffusion Models

We propose a Likelihood Matching approach for training diffusion models by first establishing an equivalence between the likelihood of the target data distribution and a likelihood along the sample path of the reverse diffusion. To efficiently compute the reverse sample likelihood, a quasi-likelihood is considered to approximate each reverse transition density by a Gaussian distribution with matched conditional mean and covariance, respectively. The score and Hessian functions for the diffusion generation are estimated by maximizing the quasi-likelihood, ensuring a consistent matching of both the first two transitional moments between every two time points. A stochastic sampler is introduced to facilitate computation that leverages both the estimated score and Hessian information. We establish consistency of the quasi-maximum likelihood estimation, and provide non-asymptotic convergence guarantees for the proposed sampler, quantifying the rates of the approximation errors due to the score and Hessian estimation, dimensionality, and the number of diffusion steps. Empirical and simulation evaluations demonstrate the effectiveness of the proposed Likelihood Matching and validate the theoretical results.

stat.ML↗

arXiv2024

VascularPilot3D: Toward a 3D fully autonomous navigation for endovascular robotics

This research reports VascularPilot3D, the first 3D fully autonomous endovascular robot navigation system. As an exploration toward autonomous guidewire navigation, VascularPilot3D is developed as a complete navigation system based on intra-operative imaging systems (fluoroscopic X-ray in this study) and typical endovascular robots. VascularPilot3D adopts previously researched fast 3D-2D vessel registration algorithms and guidewire segmentation methods as its perception modules. We additionally propose three modules: a topology-constrained 2D-3D instrument end-point lifting method, a tree-based fast path planning algorithm, and a prior-free endovascular navigation strategy. VascularPilot3D is compatible with most mainstream endovascular robots. Ex-vivo experiments validate that VascularPilot3D achieves 100% success rate among 25 trials. It reduces the human surgeon's overall control loops by 18.38%. VascularPilot3D is promising for general clinical autonomous endovascular navigations.

cs.RO↗

arXiv2023

FedDBL: Communication and Data Efficient Federated Deep-Broad Learning for Histopathological Tissue Classification

Histopathological tissue classification is a fundamental task in computational pathology. Deep learning-based models have achieved superior performance but centralized training with data centralization suffers from the privacy leakage problem. Federated learning (FL) can safeguard privacy by keeping training samples locally, but existing FL-based frameworks require a large number of well-annotated training samples and numerous rounds of communication which hinder their practicability in the real-world clinical scenario. In this paper, we propose a universal and lightweight federated learning framework, named Federated Deep-Broad Learning (FedDBL), to achieve superior classification performance with limited training samples and only one-round communication. By simply associating a pre-trained deep learning feature extractor, a fast and lightweight broad learning inference system and a classical federated aggregation approach, FedDBL can dramatically reduce data dependency and improve communication efficiency. Five-fold cross-validation demonstrates that FedDBL greatly outperforms the competitors with only one-round communication and limited training samples, while it even achieves comparable performance with the ones under multiple-round communications. Furthermore, due to the lightweight design and one-round communication, FedDBL reduces the communication burden from 4.6GB to only 276.5KB per client using the ResNet-50 backbone at 50-round training. Since no data or deep model sharing across different clients, the privacy issue is well-solved and the model security is guaranteed with no model inversion attack risk. Code is available at https://github.com/tianpeng-deng/FedDBL.

eess.IV↗

arXiv2022

Ultra-high energy cosmic neutrinos from gamma-ray bursts

Based on recent proposal to associate IceCube TeV and PeV neutrino events with gamma-ray bursts~(GRBs) by considering the Lorentz violation of neutrinos, we provide a new estimate on the GRB neutrino flux and such result is much bigger than previous results by the IceCube Collaboration. Among these 24 neutrino ``shower" events above 60~TeV, 12 events are associated with GRBs. Such result is comparable with the prediction from GRB fireball models. Analysis of track events provide consistent result with the shower events to associate high energy cosmic neutrinos with GRBs under the same Lorentz violation features of neutrinos. We also make a background estimation and reveal GRBs as a significant source for the ultra-high energy IceCube neutrino events. Our work supports the Lorentz violation and $CPT$-violation of neutrinos, indicating new physics beyond relativity.

astro-ph.HE↗

arXiv2020

Experimental and numerical studies on kV scattered x-ray imaging for real-time image guidance in radiation therapy

Motion management is a critical component of image guidance radiotherapy for lung cancer. We previously proposed a scheme using kV scattered x-ray photons for marker-less real-time image guidance in lung cancer radiotherapy. This study reports our recently progress using the photon counting detection technique to demonstrate potential feasibility of this method and using Monte Carlo (MC) simulations and ray-tracing calculations to characterize the performance. In our scheme, a thin slice of x-ray beam was directed to the target and we measured the outgoing scattered photons using a photon counting detector with a parallel-hole collimator to establish the correspondence between detector pixels and scatter positions. Image corrections of geometry, beam attenuation and scattering angle were performed to convert the raw image to the actual image of Compton attenuation coefficient. We set up a MC simulation system using an in-house developed GPU-based MC package modeling the image formation process. We also performed ray-tracing calculations to investigate the impacts of imaging system geometry on resulting image resolution. The experiment demonstrated feasibility of using a photon counting detector to measure scattered x-ray photons and generate the proposed scattered x-ray image. After correction, x-ray scattering image intensity and Compton scattering attenuation coefficient were linearly related, with R2=0.91. Contrast to Noise Ratios of different objects were improved and the values in experimental results and MC simulation results agreed with each other. Ray-tracing calculations revealed the dependence of image resolution on imaging geometry. The image resolution increases with reduced source to object distance and increased collimator height. The study demonstrated potential feasibility of using scattered x-ray imaging as a real-time image guidance method in radiation therapy.

physics.med-ph↗

arXiv2019

Consistent Lorentz violation features from near-TeV IceCube neutrinos

A recent proposal to associate 60~TeV to 2~PeV IceCube neutrino events with gamma-ray bursts~(GRBs) indicates the Lorentz violation of cosmic neutrinos and leads further to the $CPT$ symmetry violation between neutrinos and antineutrinos. Here we find that another 12 northern hemisphere track events possibly correlated with GRBs from three-year IceCube data satisfy the same regularity at a lower energy scale around 1~TeV. The combined fitting indicates a Lorentz violation scale ${E}_{\rm LV}=(6.4\pm 1.5)\times10^{17}~{ \rm GeV}$ and an intrinsic time difference ${Δ{t}_{\rm in}=(-2.8\pm 0.7)\times10^2~{\rm s}}$, from which we find an earlier emission of neutrinos than photons at the GRB source. We also suggest analyzing neutrino events detected a few minutes before the GRB trigger time to test the $CPT$ violation of ultrahigh-energy neutrinos.

hep-ph↗

arXiv2018

Lorentz violation from gamma-ray burst neutrinos

The Lorentz violation~(LV) effect of ultra-relativistic particles can be tested by gamma-ray burst~(GRB) neutrinos and photons. The IceCube Collaboration has observed plenty of ultra-high energy neutrinos, including four events of PeV scale neutrinos. Recent studies suggested a possible energy dependent speed variation of GRB neutrinos in a similar way to that of GRB photons. Here we find that all four events of PeV neutrinos with associated GRB candidates can satisfy a regularity found from TeV neutrinos about a linear form correlation between the observed time difference and the LV factor. Such regularity indicates a Lorentz violation scale $E_{\rm LV}=(6.5\pm 0.4)\times10^{17}~{\rm GeV}$, which is comparable with that determined by GRB photons. We also suggest that neutrinos and anti-neutrinos can be superluminal and subluminal respectively due to opposite signs of LV correction.

hep-ph↗

arXiv2018

Quantifying Quantum Coherence in Experimentally-Observed Neutrino Oscillations

Neutrino oscillation represents an intriguing physical phenomenon where the quantumness can be maintained and detected over a long distance. Previously, the non-classical character of neutrino oscillation was tested with the Leggett-Garg inequality, where a clear violation of the classical bound was observed [J. A. Formaggio et al., Phys. Rev. Lett. 117, 050402 (2016)]. However, there are several limitations in testing neutrino oscillations with the Leggett-Garg inequality. In particular, the degree of violation of the Leggett-Garg inequality cannot be taken as a "measure of quantumness". Here we focus on quantifying the quantumness of experimentally-observed neutrino oscillation, using the tools of recently-developed quantum resource theory. We analyzed ensembles of reactor and accelerator neutrinos at distinct energies from a variety of neutrino sources, including Daya Bay (0.5 km and 1.6 km), Kamland (180 km), MINOS (735 km), and T2K (295 km). The quantumness of the three-flavoured neutrino oscillation is characterized within a 3σ range relative to the theoretical prediction. It is found that the maximal coherence was observed in the neutrino source from the Kamland reactor. However, even though the survival probability of the Daya Bay experiment did not vary significantly (dropped about 10 percent), the coherence recorded can reach up to 40 percent of the maximal value. These results represent the longest distance over which quantumness were experimentally determined for quantum particles other than photons.

hep-ph↗

arXiv2016

Neutrino properties from ultra-high energy cosmic neutrinos

Neutrino properties can be constrained by the detection of ultra-high energy cosmic neutrinos (UHECNs). By using the updated global fitting results of neutrino mixing parameters, we present predictions on the neutrino flavor ratios at the Earth from three possibly astrophysical sources. Comparing with the latest IceCube data, we find that the normal hierarchy (NH) and inverted hierarchy (IH) cases from the initial ratios $ϕ_{ν_e}^0:ϕ_{ν_μ}^0:ϕ_{ν_τ}^0=$1:2:0 and 0:1:0 are compatible with the data in the standard neutrino oscillation scenario. We also examine the neutrino flavor ratios in a neutrino decay scenario beyond the standard model, and introduce the special case that two mass eigenstates of neutrinos, i.e., $ν_1$ and $ν_2$, are degenerated. We find that the IH case and the degenerate NH case from the 1:2:0 and 0:1:0 sources are still permissible with the IceCube data within the $3σ$ error range. The general constraints only rely on the neutrino mixing and oscillation framework are also discussed.

hep-ph↗

arXiv2014

Constraints on absolute neutrino Majorana mass from current data

We present new constraints on the neutrino Majorana masses from the current data of neutrinoless double beta decay and neutrino flavour mixing. With the latest results of $0νββ$ progresses from various isotopes, including the recent calculations of the nuclear matrix elements, we find that the strongest constraint of the effective Majorana neutrino mass is from the $^{136}\rm{Xe}$ data of the EXO-200 and KamLAND-Zen collaborations. Further more, by combining the $0νββ$ experimental data with the neutrino mixing parameters from new analyses, we get the mass upper limits of neutrino mass eigenstates and flavour eigenstates and suggest several relations among these neutrino masses.

hep-ph↗