arXiv subjects
Publications and source records attributed to Xiaochun Li.
We prove the boundedness of multilinear Hilbert transforms along moment curves by establishing uniform estimates for both translated and untranslated multilinear paraproducts and by applying a Sobolev smoothing inequality. As a byproduct of our method, we obtain the same bounds for multilinear maximal functions along moment curves.
We demonstrate the almost everywhere convergence of the planar Bochner-Riesz means for $L^p$ functions in the optimal range when $5/3\leq p\leq 2$. This is achieved by establishing a sharp $L^{5/3}$ estimate for a maximal operator closely associated with the Bochner-Riesz multiplier operator. The estimate depends on a new refined $L^2$ estimate, which may be of independent interest.
We prove sharp local smoothing estimates for wave equations on compact Riemannian manifolds in $n+1$ dimensions for odd $n$ and obtain improved estimates in even dimensions. This is achieved by deriving local smoothing estimates for certain Fourier integral operators. We also obtain improved local smoothing estimates for wave equations in Euclidean spaces.
In the early stages of semiconductor equipment development, obtaining large quantities of raw optical images poses a significant challenge. This data scarcity hinder the advancement of AI-powered solutions in semiconductor manufacturing. To address this challenge, we introduce SinSEMI, a novel one-shot learning approach that generates diverse and highly realistic images from single optical image. SinSEMI employs a multi-scale flow-based model enhanced with LPIPS (Learned Perceptual Image Patch Similarity) energy guidance during sampling, ensuring both perceptual realism and output variety. We also introduce a comprehensive evaluation framework tailored for this application, which enables a thorough assessment using just two reference images. Through the evaluation against multiple one-shot generation techniques, we demonstrate SinSEMI's superior performance in visual quality, quantitative measures, and downstream tasks. Our experimental results demonstrate that SinSEMI-generated images achieve both high fidelity and meaningful diversity, making them suitable as training data for semiconductor AI applications.
For $p\ge 2$, and $λ>\max\{n|\tfrac 1p-\tfrac 12|-\tfrac12, 0\}$, we prove the pointwise convergence of cone multipliers, i.e. $$ \lim_{t\to\infty}T_t^λ(f)\to f \text{ a.e.},$$ where $f\in L^p(\mathbb R^n)$ satisfies $supp\ \widehat f\subset\{ξ\in\mathbb R^n:\ 1<|ξ_n|<2\}$. Our main tools are weighted estimates for maximal cone operators, which are consequences of trace inequalities for cones.
A Stein-Tomas type inequality and a (weak) decoupling inequality are proved by using the polynomial partitioning method. Both estimates are related closely to Waring's problem.
Assuming some pointwise estimates on certain Weyl's sum, we prove the sharp estimates of the mean value associated to the following exponential sum $$ \sum_{n=1}^N e^{2πi tn^d +2πi xn}\,. $$
Using Bombieri-Iwaniec method, we establish an improvement on both Gauss's Circle Problem and Dirichlet's Divisor Problem. More precisely, we derive a new estimate for the first spacing problem and combine it with Huxley's results on the second spacing problem.
Lithium metal has been an attractive candidate as a next generation anode material. Despite its popularity, stability issues of lithium in the liquid electrolyte and the formation of lithium whiskers have kept it from practical use. Three-dimensional (3D) current collectors have been proposed as an effective method to mitigate whiskers growth. Although extensive research efforts have been done, the effects of three key parameters of the 3D current collectors, namely the surface area, the tortuosity factor, and the surface chemistry, on the performance of lithium metal batteries remain elusive. Herein, we quantitatively studied the role of these three parameters by synthesizing four types of porous copper networks with different sizes of well-structured micro-channels. X-ray microscale computed tomography (micro-CT) allowed us to assess the surface area, the pore size and the tortuosity factor of the porous copper materials. A metallic Zn coating was also applied to study the influence of surface chemistry on the performance of the 3D current collectors. The effects of these parameters on the performance were studied in detail through Scanning Electron Microscopy (SEM) and Titration Gas Chromatography (TGC). Stochastic simulations further allowed us to interpret the role of the tortuosity factor in lithiation. By understanding these effects, the optimal range of the key parameters is found for the porous copper anodes and their performance is predicted. Using these parameters to inform the design of porous copper anodes for Li deposition, Coulombic efficiencies (CE) of up to 99.56% are achieved, thus paving the way for the design of effective 3D current collector systems.
We prove some new $L^p$ estimates for maximal Bochner-Riesz operator in the plane.
We give two Roth theorems, related to the nonlinear configuration $x$, $x+P_1(t)$, $x+P_2(t)$ involving two polynomials, for sets in $\mathbb{R}$ of positive density and of fractional dimensions. The proof uses Fourier analysis.
Image retrieval utilizes image descriptors to retrieve the most similar images to a given query image. Convolutional neural network (CNN) is becoming the dominant approach to extract image descriptors for image retrieval. For low-power hardware implementation of image retrieval, the drawback of CNN-based feature descriptor is that it requires hundreds of megabytes of storage. To address this problem, this paper applies deep model quantization and compression to CNN in ASIC chip for image retrieval. It is demonstrated that the CNN-based features descriptor can be extracted using as few as 2-bit weights quantization to deliver a similar performance as floating-point model for image retrieval. In addition, to implement CNN in ASIC, especially for large scale images, the limited buffer size of chips should be considered. To retrieve large scale images, we propose an improved pooling strategy, region nested invariance pooling (RNIP), which uses cropped sub-images for CNN. Testing results on chip show that integrating RNIP with the proposed 2-bit CNN model compression approach is capable of retrieving large scale images.
Phase III randomized clinical trials play a monumentally critical role in the evaluation of new medical products. Because of the intrinsic nature of uncertainty embedded in our capability in assessing the efficacy of a medical product, interpretation of trial results relies on statistical principles to control the error of false positives below desirable level. The well-established statistical hypothesis testing procedure suffers from two major limitations, namely, the lack of flexibility in the thresholds to claim success and the lack of capability of controlling the total number of false positives that could be yielded by the large volume of trials. We propose two general theoretical frameworks based on the conventional frequentist paradigm and Bayesian perspectives, which offer realistic, flexible and effective solutions to these limitations. Our methods are based on the distribution of the effect sizes of the population of trials of interest. The estimation of this distribution is practically feasible as clinicaltrials.gov provides a centralized data repository with unbiased coverage of clinical trials. We provide a detailed development of the two frameworks with numerical results obtained for industry sponsored Phase III randomized clinical trials.
We obtain partial improvement toward the pointwise convergence problem of Schrödinger solutions, in the general setting of fractal measure. In particular, we show that, for $n\geq 3$, $\lim_{t \to 0} e^{itΔ}f(x) = f(x)$ almost everywhere with respect to Lebesgue measure for all $f \in H^s (\mathbb{R}^n)$ provided that $s>(n+1)/2(n+2)$. The proof uses linear refined Strichartz estimates. We also prove a multilinear refined Strichartz using decoupling and multilinear Kakeya.
To prove Fourier restriction estimate using polynomial partitioning, Guth introduced the concept of $k$-broad part of regular $L^p$ norm and obtained sharp $k$-broad restriction estimates. To go from $k$-broad estimates to regular $L^p$ estimates, Guth employed $l^2$ decoupling result. In this article, similar to the technique introduced by Bourgain-Guth, we establish an analogue to go from regular $L^p$ norm to its $(m+1)$-broad part, as the error terms we have the restricted $k$-broad parts ($k=2,\cdots,m$). To analyze the restricted $k$-broadness, we prove an $l^p$ decoupling result, which can be applied to handle the error terms and recover Guth's linear restriction estimates.
We prove that, under certain conditions on the function pair $φ_1$ and $φ_2$, bilinear average $p^{-1}\sum_{y\in \mathbb{F}_p}f_1(x+φ_1(y)) f_2(x+φ_2(y))$ along curve $(φ_1, φ_2)$ satisfies certain decay estimate. As a consequence, Roth type theorems hold in the setting of finite fields. In particular, if $φ_1,φ_2\in \mathbb{F}_p[X]$ with $φ_1(0)=φ_2(0)=0$ are linearly independent polynomials, then for any $A\subset \mathbb{F}_p, |A|=δp$ with $δ>c p^{-\frac{1}{12}}$, there are $\gtrsim δ^3p^2$ triplets $x,x+φ_1(y), x+φ_2(y)\in A$. This extends a recent result of Bourgain and Chang who initiated this type of problems, and strengthens the bound in a result of Peluse, who generalized Bourgain and Chang's work. The proof uses discrete Fourier analysis and algebraic geometry.
We show that $\lim_{t \to 0} e^{itΔ}f(x) = f(x)$ almost everywhere for all $f \in H^s (\mathbb{R}^2)$ provided that $s>1/3$. This result is sharp up to the endpoint. The proof uses polynomial partitioning and decoupling.
Using Guth's polynomial partitioning method, we obtain $L^p$ estimates for the maximal function associated to the solution of Schrödinger equation in $\mathbb R^2$. The $L^p$ estimates can be used to recover the previous best known result that $\lim_{t \to 0} e^{itΔ}f(x)=f(x)$ almost everywhere for all $f \in H^s (\mathbb{R}^2)$ provided that $s>3/8$.