Fei Lu
YOU?
Author Swipe
View article: Dimension-Free Minimax Rates for Learning Pairwise Interactions in Attention-Style Models
Dimension-Free Minimax Rates for Learning Pairwise Interactions in Attention-Style Models Open
We study the convergence rate of learning pairwise interactions in single-layer attention-style models, where tokens interact through a weight matrix and a non-linear activation function. We prove that the minimax rate is $M^{-\frac{2β}{2β…
View article: An Adaptive RKHS Regularization for the Fredholm Integral Equations
An Adaptive RKHS Regularization for the Fredholm Integral Equations Open
Regularization is a long‐standing challenge for ill‐posed linear inverse problems, and a prototype is the Fredholm integral equation of the first kind with additive Gaussian measurement noise. We regularize by a new reproducing kernel Hilb…
View article: Genome surveillance of SARS-CoV-2 variants in Wuhan in June 2023
Genome surveillance of SARS-CoV-2 variants in Wuhan in June 2023 Open
The COVID-19 pandemic caused by SARS-CoV-2 has had a significant impact on global social and economic activities. The continuous emergence of new SARS-CoV-2 variants has made the global pandemic of COVID-19 difficult to predict. Therefore,…
View article: Cubic bricks that every b-invariant edge is forcing
Cubic bricks that every b-invariant edge is forcing Open
A connected graph G is matching covered if every edge lies in some perfect matching of G. Lovasz proved that every matching covered graph G can be uniquely decomposed into a list of bricks (nonbipartite) and braces (bipartite) up to multip…
View article: Planar wheel-like bricks
Planar wheel-like bricks Open
An edge e in a matching covered graph G is removable if G-e is matching covered; a pair {e; f} of edges of G is a removable doubleton if G-e-f is matching covered, but neither G-e nor G-f is. Removable edges and removable doubletons are ca…
View article: Robust First and Second-Order Differentiation for Regularized Optimal Transport
Robust First and Second-Order Differentiation for Regularized Optimal Transport Open
Applications such as unbalanced and fully shuffled regression can be approached by optimizing regularized optimal transport (OT) distances, such as the entropic OT and Sinkhorn distances. A common approach for this optimization is to use a…
View article: RHOF activation of AKT/β-catenin signaling pathway drives acute myeloid leukemia progression and chemotherapy resistance
RHOF activation of AKT/β-catenin signaling pathway drives acute myeloid leukemia progression and chemotherapy resistance Open
Acute myeloid leukemia (AML) is a clonal malignancy originating from leukemia stem cells, characterized by a poor prognosis, underscoring the necessity for novel therapeutic targets and treatment methodologies. This study focuses on Ras ho…
View article: Probabilistic cellular automata with local transition matrices: synchronization, ergodicity, and inference
Probabilistic cellular automata with local transition matrices: synchronization, ergodicity, and inference Open
We introduce a new class of probabilistic cellular automata that are capable of exhibiting rich dynamics such as synchronization and ergodicity and can be easily inferred from data. The system is a finite-state locally interacting Markov c…
View article: Interacting Particle Systems on Networks: joint inference of the network and the interaction kernel
Interacting Particle Systems on Networks: joint inference of the network and the interaction kernel Open
Modeling multi-agent systems on networks is a fundamental challenge in a wide variety of disciplines. We jointly infer the weight matrix of the network and the interaction kernel, which determine respectively which agents interact with whi…
View article: Alternative modified Cholesky decomposition of the precision matrix of longitudinal data
Alternative modified Cholesky decomposition of the precision matrix of longitudinal data Open
The correlation matrix might be of scientific interest for longitudinal data. However, few studies have focused on both robust estimation of the correlation matrix against model misspecification and robustness to outliers in the data, when…
View article: Optimal minimax rate of learning nonlocal interaction kernels
Optimal minimax rate of learning nonlocal interaction kernels Open
Nonparametric estimation of nonlocal interaction kernels is crucial in various applications involving interacting particle systems. The inference challenge, situated at the nexus of statistical learning and inverse problems, arises from th…
View article: Energy-Aware Routing Algorithm for Mobile Ground-to-Air Charging
Energy-Aware Routing Algorithm for Mobile Ground-to-Air Charging Open
We investigate the problem of energy-constrained planning for a cooperative system of an Unmanned Ground Vehicles (UGV) and an Unmanned Aerial Vehicle (UAV). In scenarios where the UGV serves as a mobile base to ferry the UAV and as a char…
View article: Ginsenoside protects intestinal barrier function and improves epithelium injury in sepsis by regulating the miR-30e-5p/FBXO11 axis
Ginsenoside protects intestinal barrier function and improves epithelium injury in sepsis by regulating the miR-30e-5p/FBXO11 axis Open
Background: Ginsenoside Rg1 has been studied to improve systemic inflammatory injury induced by sepsis, but its mechanism is not fully understood. The objective of this study was to explore the potential molecular mechanism by which Rg1 am…
View article: Statistical Learning for Nonlinear Model Reduction from Local Simulations of Stochastic and Particle- and Agent-Based Systems
Statistical Learning for Nonlinear Model Reduction from Local Simulations of Stochastic and Particle- and Agent-Based Systems Open
Stochastic physical systems across the sciences that have very high-dimensional state spaces, with a large number of fast degrees of freedom that force direct simulators to proceed by integration steps that are orders of magnitude smaller …
View article: The $Λ$-type P-wave bottom baryon states via the QCD sum rules
The $Λ$-type P-wave bottom baryon states via the QCD sum rules Open
Our study focuses on the $Λ$-type P-wave bottom baryon states with the spin-parity $J^P=\frac{1}{2}^-$, $\frac{3}{2}^-$. We introduce an explicit P-wave between the two light quarks in the interpolating currents (and the two light quarks a…
View article: Small noise analysis for Tikhonov and RKHS regularizations
Small noise analysis for Tikhonov and RKHS regularizations Open
Regularization plays a pivotal role in ill-posed machine learning and inverse problems. However, the fundamental comparative analysis of various regularization norms remains open. We establish a small noise analysis framework to assess the…
View article: Analysis of the D-wave $Σ$-type charmed baryon states with the QCD sum rules
Analysis of the D-wave $Σ$-type charmed baryon states with the QCD sum rules Open
We construct the $Σ$-type currents to investigate the D-wave charmed baryon states with the QCD sum rules systematically. The predicted masses $M=3.35^{+0.13}_{-0.18}\,\rm{GeV}$ ($3.33^{+0.13}_{-0.16}\,\rm{GeV}$), $3.34^{+0.14}_{-0.18}\,\r…
View article: An adaptive RKHS regularization for Fredholm integral equations
An adaptive RKHS regularization for Fredholm integral equations Open
Regularization is a long-standing challenge for ill-posed linear inverse problems, and a prototype is the Fredholm integral equation of the first kind with additive Gaussian measurement noise. We introduce a new RKHS regularization adaptiv…
View article: Comment on essd-2022-428
Comment on essd-2022-428 Open
Abstract. Surface ozone is an important air pollutant detrimental to human health and vegetation productivity. Regardless of its short atmospheric lifetime, surface ozone has significantly increased since the 1970s across the Northern Hemi…
View article: Benchmarking optimality of time series classification methods in distinguishing diffusions
Benchmarking optimality of time series classification methods in distinguishing diffusions Open
Statistical optimality benchmarking is crucial for analyzing and designing time series classification (TSC) algorithms. This study proposes to benchmark the optimality of TSC algorithms in distinguishing diffusion processes by the likeliho…
View article: Reply on CC1
Reply on CC1 Open
Abstract. Surface ozone is an important air pollutant detrimental to human health and vegetation productivity. Regardless of its short atmospheric lifetime, surface ozone has significantly increased since the 1970s across the Northern Hemi…
View article: Unsupervised learning of observation functions in state space models by nonparametric moment methods
Unsupervised learning of observation functions in state space models by nonparametric moment methods Open
We investigate the unsupervised learning of non-invertible observation functions in nonlinear state space models. Assuming abundant data of the observation process along with the distribution of the state process, we introduce a nonparamet…
View article: Identifiability of interaction kernels in mean-field equations of interacting particles
Identifiability of interaction kernels in mean-field equations of interacting particles Open
This study examines the identifiability of interaction kernels in mean-field equations of interacting particles or agents, an area of growing interest across various scientific and engineering fields. The main focus is identifying data-dep…
View article: A Data-Adaptive Prior for Bayesian Learning of Kernels in Operators
A Data-Adaptive Prior for Bayesian Learning of Kernels in Operators Open
Kernels are efficient in representing nonlocal dependence and they are widely used to design operators between function spaces. Thus, learning kernels in operators from data is an inverse problem of general interest. Due to the nonlocal de…
View article: Differential cross-section measurements for neutron-induced $α$ production reactions on carbon across neutron energy range of 6.2 to 76 MeV
Differential cross-section measurements for neutron-induced $α$ production reactions on carbon across neutron energy range of 6.2 to 76 MeV Open
Angle-differential cross sections for neutron-induced $α$ production in carbon were determined at thirty discrete neutron energy levels ranging from 6.2 to 76 MeV at the Back-n white neutron source of the China Spallation Neutron Source. U…
View article: Stochastic Data-Driven Variational Multiscale Reduced Order Models
Stochastic Data-Driven Variational Multiscale Reduced Order Models Open
Trajectory-wise data-driven reduced order models (ROMs) tend to be sensitive to training data, and thus lack robustness. We propose to construct a robust stochastic ROM closure (S-ROM) from data consisting of multiple trajectories from ran…