Source author record

Mingwei Zhang

Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.

ResearcherUnclaimed source record

eess.AS Machine Learning math-ph math.AG math.AP math.CO math.DG math.MP Sound

Catalog footprint

Research overview

4works

9topics

4close collaborators

Actions

Is this your work?

Open graph Browse papers

preprint2026arXiv

Conformal invariants for the zero mode equation

For non-trivial solutions to the zero mode equation on a closed spin manifold $D φ=iA\cdot φ,$ we first provide a simple proof for the sharp inequality \eq{ \norm{A}_{L^n}^2 \ge \frac {n}{4(n-1)} Y(M,[g]), } where $Y(M,[g])$ is the Yamabe constant of $(M,g)$ , which was obtained by Frank-Loss and Reuss. Then we classify completely the equality case by proving that equality holds if and only if $φ$ is a Killing spinor, and if and only if $(M,g)$ is a Sasaki-Einstein manifold with $A$ (up to scaling) as its Reeb field and $φ$ a vacuum up to a conformal transformation. More generalizations have been also studied.

preprint2020arXiv

Generalized Operating Procedure for Deep Learning: an Unconstrained Optimal Design Perspective

Deep learning (DL) has brought about remarkable breakthrough in processing images, video and speech due to its efficacy in extracting highly abstract representation and learning very complex functions. However, there is seldom operating procedure reported on how to make it for real use cases. In this paper, we intend to address this problem by presenting a generalized operating procedure for DL from the perspective of unconstrained optimal design, which is motivated by a simple intension to remove the barrier of using DL, especially for those scientists or engineers who are new but eager to use it. Our proposed procedure contains seven steps, which are project/problem statement, data collection, architecture design, initialization of parameters, defining loss function, computing optimal parameters, and inference, respectively. Following this procedure, we build a multi-stream end-to-end speaker verification system, in which the input speech utterance is processed by multiple parallel streams within different frequency range, so that the acoustic modeling can be more robust resulting from the diversity of features. Trained with VoxCeleb dataset, our experimental results verify the effectiveness of our proposed operating procedure, and also show that our multi-stream framework outperforms single-stream baseline with 20 % relative reduction in minimum decision cost function (minDCF).

preprint2016arXiv

Some extremal results on complete degenerate hypergraphs

Let $K^{(r)}_{s_1,s_2,\cdots,s_r}$ be the complete $r$ -partite $r$ -uniform hypergraph and $ex(n,K^{(r)}_{s_1,s_2,\cdots,s_r})$ be the maximum number of edges in any $n$ -vertex $K^{(r)}_{s_1,s_2,\cdots,s_r}$ -free $r$ -uniform hypergraph. It is well-known in the graph case that $ex(n,K_{s,t})=Θ(n^{2-1/s})$ when $t$ is sufficiently larger than $s$ . In this note, we generalize the above to hypergraphs by showing that if $s_r$ is sufficiently larger than $s_1,s_2,\cdots,s_{r-1}$ then $ex(n, K^{(r)}_{s_1,s_2,\cdots,s_r})=Θ\left(n^{r-\frac{1}{s_1s_2\cdots s_{r-1}}}\right).$ This follows from a more general Turán type result we establish in hypergraphs, which also improves and generalizes some recent results of Alon and Shikhelman. The lower bounds of our results are obtained by the powerful random algebraic method of Bukh. Another new, perhaps unsurprising insight which we provide here is that one can also use the random algebraic method to construct non-degenerate (hyper-)graphs for various Turán type problems. The asymptotics for $ex(n, K^{(r)}_{s_1,s_2,\cdots,s_r})$ is also proved by Verstraëte independently with a different approach.

preprint2014arXiv

On the Chern number inequalities satisfied by all smooth complete intersection threefolds with ample canonical class

We obtain all linear Chern number inequalities satisfied by any smooth complete intersection threefold with ample canonical bundle.

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.