-
Chromatic symmetric functions of conjoined graphs
Authors:
E. Y. J. Qi,
D. Q. B. Tang,
D. G. L. Wang
Abstract:
We introduce path-conjoined graphs defined for two rooted graphs by joining their roots with a path, and investigate the chromatic symmetric functions of its two generalizations: spider-conjoined graphs and chain-conjoined graphs. By using the composition method developed by Zhou and the third author, we obtain neat positive $e_I$-expansions for the chromatic symmetric functions of clique-path-cyc…
▽ More
We introduce path-conjoined graphs defined for two rooted graphs by joining their roots with a path, and investigate the chromatic symmetric functions of its two generalizations: spider-conjoined graphs and chain-conjoined graphs. By using the composition method developed by Zhou and the third author, we obtain neat positive $e_I$-expansions for the chromatic symmetric functions of clique-path-cycle graphs, path-clique-path graphs, and path-clique-clique graphs.
△ Less
Submitted 3 June, 2024;
originally announced June 2024.
-
Harmonic $1$-forms on real loci of Calabi-Yau manifolds
Authors:
Michael R. Douglas,
Daniel Platt,
Yidi Qi
Abstract:
We numerically study whether there exist nowhere vanishing harmonic $1$-forms on the real locus of some carefully constructed examples of Calabi-Yau manifolds, which would then give rise to potentially new examples of $G_2$-manifolds and an explicit description of their metrics. We do this in two steps: first, we use a neural network to compute an approximate Calabi-Yau metric on each manifold. Se…
▽ More
We numerically study whether there exist nowhere vanishing harmonic $1$-forms on the real locus of some carefully constructed examples of Calabi-Yau manifolds, which would then give rise to potentially new examples of $G_2$-manifolds and an explicit description of their metrics. We do this in two steps: first, we use a neural network to compute an approximate Calabi-Yau metric on each manifold. Second, we use another neural network to compute an approximately harmonic $1$-form with respect to the approximate metric, and then inspect the found solution. On two manifolds existence of a nowhere vanishing harmonic $1$-form can be ruled out using differential geometry. The real locus of a third manifold is diffeomorphic to $S^1 \times S^2$, and our numerics suggest that when the Calabi-Yau metric is close to a singular limit, then it admits a nowhere vanishing harmonic $1$-form. We explain how such an approximate solution could potentially be used in a numerically verified proof for the fact that our example manifold must admit a nowhere vanishing harmonic $1$-form.
△ Less
Submitted 29 May, 2024;
originally announced May 2024.
-
Eigenvalues of Product of Ginibre Ensembles and Their Inverses and that of Truncated Haar Unitary Matrices and Their Inverses
Authors:
Shuhua Chang,
Tiefeng Jiang,
Yongcheng Qi
Abstract:
Consider two types of products of independent random matrices, including products of Ginibre matrices and inverse Ginibre matrices and products of truncated Haar unitary matrices and inverse truncated Haar matrices. Each product matrix has $m$ multiplicands of $n$ by $n$ square matrices, and the empirical distribution based on the $n$ eigenvalues of the product matrix is called empirical spectral…
▽ More
Consider two types of products of independent random matrices, including products of Ginibre matrices and inverse Ginibre matrices and products of truncated Haar unitary matrices and inverse truncated Haar matrices. Each product matrix has $m$ multiplicands of $n$ by $n$ square matrices, and the empirical distribution based on the $n$ eigenvalues of the product matrix is called empirical spectral distribution of the matrix. In this paper, we investigate the limiting empirical spectral distribution of the product matrices when $n$ tends to infinity and $m$ changes with $n$. For properly rescaled eigenvalues for two types of the product matrices, we obtain the necessary and sufficient conditions for the convergence of the empirical spectral distributions.
△ Less
Submitted 12 March, 2024;
originally announced March 2024.
-
Combinatorial p-th Calabi Flows for Total Geodesic Curvatures in hyperbolic background geometry
Authors:
Guangming Hu,
Zi** Lei,
Yi Qi,
Puchun Zhou
Abstract:
In hyperbolic background geometry, we investigate a generalized circle packing (including circles, horocycles and hypercycles) with conical singularities on a surface with boundary, which has a total geodesic curvature on each generalized circle of this circle packing and a discrete Gaussian curvature on the center of each dual circle. The purpose of this paper is to find this type of circle packi…
▽ More
In hyperbolic background geometry, we investigate a generalized circle packing (including circles, horocycles and hypercycles) with conical singularities on a surface with boundary, which has a total geodesic curvature on each generalized circle of this circle packing and a discrete Gaussian curvature on the center of each dual circle. The purpose of this paper is to find this type of circle packings with prescribed total geodesic curvatures on generalized circles and discrete Gaussian curvatures on centers of dual circles. To achieve this goal, we firstly establish existence and rigidity on this type of circle packings by the variational principle. Secondly, for $p>1$, we introduce combinatorial $p$-th Calabi flows to find the circle packing with prescribed total geodesic curvatures on generalized circles and discrete Gaussian curvatures on centers of dual circles for the first time.
△ Less
Submitted 8 March, 2024;
originally announced March 2024.
-
Bipartitions with prescribed order of highly connected digraphs
Authors:
Yuzhen Qi,
** Yan,
Jia Zhou
Abstract:
A digraph is strongly connected if it has a directed path from $x$ to $y$ for every ordered pair of distinct vertices $x, y$ and it is strongly $k$-connected if it has at least $k+1$ vertices and remains strongly connected when we delete any set of at most $k-1$ vertices. For a digraph $D$, we use $δ(D)$ to denote $\mathop{\text{min}}\limits_{v\in V (D)} {|N_D^+(v)\cup N_D^-(v)|}$. In this paper,…
▽ More
A digraph is strongly connected if it has a directed path from $x$ to $y$ for every ordered pair of distinct vertices $x, y$ and it is strongly $k$-connected if it has at least $k+1$ vertices and remains strongly connected when we delete any set of at most $k-1$ vertices. For a digraph $D$, we use $δ(D)$ to denote $\mathop{\text{min}}\limits_{v\in V (D)} {|N_D^+(v)\cup N_D^-(v)|}$. In this paper, we show the following result. Let $k, l, n, n_1, n_2 \in \mathbb{N}$ with $n_1+n_2\leq n$ and $n_1,n_2\geq n/20$. Suppose that $D$ is a strongly $10^7k(k+l)^2\log(2kl$)-connected digraph of order $n$ with $δ(D)\geq n-l$. Then there exist two disjoint subsets $V_1, V_2\in V(D)$ with $|V_1| = n_1$ and $|V_2| = n_2$ such that each of $D[V_1]$, $D[V_2]$, and $D[V_1, V_2]$ is strongly $k$-connected. In particular, $V_1$ and $V_2$ form a partition of $V(D)$ when $n_1+n_2=n$. This result improves the earlier result of Kim, Kühn, and Osthus [SIAM J. Discrete Math. 30 (2016) 895--911].
△ Less
Submitted 26 February, 2024;
originally announced February 2024.
-
Removal paths avoiding vertices
Authors:
Yuzhen Qi,
** Yan
Abstract:
In this paper, we show that for any positive integer $m$ and $k\in [2]$, let $G$ be a $(2m+2k+2)$-connected graph and let $a_1,\ldots , a_m, s, t$ be any distinct vertices of $G$, there are $k$ internally disjoint $s$-$t$ paths $P_1, \ldots, P_k$ in $G$ such that $\{a_1,\ldots , a_m\} \cap \bigcup^{k}_{i=1}V (P_i) = \emptyset$ and $G- \bigcup^{k}_{i=1}V (P_i)$ is 2-connected, which generalizes the…
▽ More
In this paper, we show that for any positive integer $m$ and $k\in [2]$, let $G$ be a $(2m+2k+2)$-connected graph and let $a_1,\ldots , a_m, s, t$ be any distinct vertices of $G$, there are $k$ internally disjoint $s$-$t$ paths $P_1, \ldots, P_k$ in $G$ such that $\{a_1,\ldots , a_m\} \cap \bigcup^{k}_{i=1}V (P_i) = \emptyset$ and $G- \bigcup^{k}_{i=1}V (P_i)$ is 2-connected, which generalizes the result by Chen, Gould and Yu [Combinatorica 23 (2003) 185--203], and Kriesell [J. Graph Theory 36 (2001) 52--58]. The case $k=1$ implies that for any $(2m+5)$-connected graph $G$, any edge $e \in E(G)$, and any distinct vertices $a_1,\ldots , a_m$ of $G-V(e)$, there exists a cycle $C$ in $G- \{a_1,\ldots , a_m\}$ such that $e\in E(C)$ and $G- V(C)$ is 2-connected, which improves the bound $10m+11$ of Y. Hong, L. Kang and X. Yu in [J. Graph Theory 80 (2015) 253--267].
△ Less
Submitted 19 February, 2024;
originally announced February 2024.
-
Reinforcement Learning from Statistical Feedback: the Journey from AB Testing to ANT Testing
Authors:
Feiyang Han,
Yimin Wei,
Zhaofeng Liu,
Yanxing Qi
Abstract:
Reinforcement Learning from Human Feedback (RLHF) has played a crucial role in the success of large models such as ChatGPT. RLHF is a reinforcement learning framework which combines human feedback to improve learning effectiveness and performance. However, obtaining preferences feedback manually is quite expensive in commercial applications. Some statistical commercial indicators are usually more…
▽ More
Reinforcement Learning from Human Feedback (RLHF) has played a crucial role in the success of large models such as ChatGPT. RLHF is a reinforcement learning framework which combines human feedback to improve learning effectiveness and performance. However, obtaining preferences feedback manually is quite expensive in commercial applications. Some statistical commercial indicators are usually more valuable and always ignored in RLHF. There exists a gap between commercial target and model training. In our research, we will attempt to fill this gap with statistical business feedback instead of human feedback, using AB testing which is a well-established statistical method. Reinforcement Learning from Statistical Feedback (RLSF) based on AB testing is proposed. Statistical inference methods are used to obtain preferences for training the reward network, which fine-tunes the pre-trained model in reinforcement learning framework, achieving greater business value. Furthermore, we extend AB testing with double selections at a single time-point to ANT testing with multiple selections at different feedback time points. Moreover, we design numerical experiences to validate the effectiveness of our algorithm framework.
△ Less
Submitted 24 November, 2023;
originally announced November 2023.
-
Hyperbolic Circle Packings and Total Geodesic Curvatures on Surfaces with Boundary
Authors:
Guangming Hu,
Yi Qi,
Yu Sun,
Puchun Zhou
Abstract:
This paper investigates a generalized hyperbolic circle packing (including circles, horocycles or hypercycles) with respect to the total geodesic curvatures on the surface with boundary. We mainly focus on the existence and rigidity of circle packing whose contact graph is the $1$-skeleton of a finite polygonal cellular decomposition, which is analogous to the construction of Bobenko and Springbor…
▽ More
This paper investigates a generalized hyperbolic circle packing (including circles, horocycles or hypercycles) with respect to the total geodesic curvatures on the surface with boundary. We mainly focus on the existence and rigidity of circle packing whose contact graph is the $1$-skeleton of a finite polygonal cellular decomposition, which is analogous to the construction of Bobenko and Springborn [4]. Motivated by Colin de Verdière's method [6], we introduce the variational principle for generalized hyperbolic circle packings on polygons. By analyzing limit behaviours of generalized circle packings on polygons, we give an existence and rigidity for the generalized hyperbolic circle packing with conical singularities regarding the total geodesic curvature on each vertex of the contact graph. As a consequence, we introduce the combinatoral Ricci flow to find a desired circle packing with a prescribed total geodesic curvature on each vertex of the contact graph.
△ Less
Submitted 17 November, 2023;
originally announced November 2023.
-
Research and experimental verification on low-frequency long-range underwater sound propagation dispersion characteristics under dual-channel sound speed profiles in the Chukchi Plateau
Authors:
**bao Weng,
Yubo Qi,
Yanming Yang,
Hongtao Wen,
Hongtao Zhou,
Ruichao Xue
Abstract:
The dual-channel sound speed profiles of the Chukchi Plateau and the Canadian Basin have become current research hotspots due to their excellent low-frequency sound signal propagation ability. Previous research has mainly focused on using sound propagation theory to explain the changes in sound signal energy. This article is mainly based on the theory of normal modes to study the fine structure of…
▽ More
The dual-channel sound speed profiles of the Chukchi Plateau and the Canadian Basin have become current research hotspots due to their excellent low-frequency sound signal propagation ability. Previous research has mainly focused on using sound propagation theory to explain the changes in sound signal energy. This article is mainly based on the theory of normal modes to study the fine structure of low-frequency wide-band sound propagation dispersion under dual-channel sound speed profiles. In this paper, the problem of the intersection of normal mode dispersion curves caused by the dual-channel sound speed profile (SSP) has been explained, the blocking effect of seabed terrain changes on dispersion structures has been analyzed, and the normal modes has been separated by using modified war** operator. The above research results have been verified through a long-range seismic exploration experiment at the Chukchi Plateau. At the same time, based on the acoustic signal characteristics in this environment, two methods for estimating the distance of sound sources have been proposed, and the experiment data at sea has also verified these two methods.
△ Less
Submitted 13 November, 2023;
originally announced November 2023.
-
Research and experimental verification on low-frequency long-range sound propagation characteristics under ice-covered and range-dependent marine environment in the Arctic
Authors:
**bao Weng,
Yubo Qi,
Yanming Yang,
Hongtao Wen,
Hongtao Zhou,
Ruichao Xue
Abstract:
At present, research on sound propagation under the Arctic ice mainly focuses on modeling and experimental verification of sound propagation under sea ice cover and unique sound velocity profiles. Among them, the main research object of concern is sound transmission loss, and this article will delve into the time-domain waveform and fine dispersion structure of low-frequency broadband acoustic sig…
▽ More
At present, research on sound propagation under the Arctic ice mainly focuses on modeling and experimental verification of sound propagation under sea ice cover and unique sound velocity profiles. Among them, the main research object of concern is sound transmission loss, and this article will delve into the time-domain waveform and fine dispersion structure of low-frequency broadband acoustic signals. Firstly, based on the theory of normal modes, this article derives the horizontal wavenumber expression and war** transformation operator for refractive normal modes in the Arctic deep-sea environment. Subsequently, based on measured ocean environmental parameters and sound field simulation calculations, this article studied the general laws of low-frequency long-range sound propagation signals in the Arctic deep-sea environment, and elucidated the impact mechanism of environmental factors such as seabed terrain changes, horizontal changes in sound velocity profiles (SSPs), and sea ice cover on low-frequency long-range sound propagation in the Arctic. This article validates the above research viewpoint through a sound propagation experiment conducted in the Arctic with a propagation distance exceeding 1000km. The marine environment of this experiment has obvious horizontal variation characteristics. At the same time, this article takes the lead in utilizing the war** transformation of refractive normal waves in the Arctic waters to achieve single hydrophone based separation of normal waves and extraction of dispersion structures, which is conducive to future research on underwater sound source localization and environmental parameter inversion based on dispersion structures.
△ Less
Submitted 13 November, 2023;
originally announced November 2023.
-
On the minimal algebraic complexity of the rank-one approximation problem for general inner products
Authors:
Khazhgali Kozhasov,
Alan Muniz,
Yang Qi,
Luca Sodomaco
Abstract:
We study the algebraic complexity of Euclidean distance minimization from a generic tensor to a variety of rank-one tensors. The Euclidean Distance (ED) degree of the Segre-Veronese variety counts the number of complex critical points of this optimization problem. We regard this invariant as a function of inner products and conjecture that it achieves its minimal value at Frobenius inner product.…
▽ More
We study the algebraic complexity of Euclidean distance minimization from a generic tensor to a variety of rank-one tensors. The Euclidean Distance (ED) degree of the Segre-Veronese variety counts the number of complex critical points of this optimization problem. We regard this invariant as a function of inner products and conjecture that it achieves its minimal value at Frobenius inner product. We prove our conjecture in the case of matrices. We discuss the above optimization problem for other algebraic varieties, classifying all possible values of the ED degree. Our approach combines tools from Singularity Theory, Morse Theory, and Algebraic Geometry.
△ Less
Submitted 26 September, 2023;
originally announced September 2023.
-
Graph Reconstruction from Connected Triples
Authors:
Yaxin Qi
Abstract:
The problem of graph reconstruction has been studied in its various forms over the years. In particular, the Reconstruction Conjecture, proposed by Ulam and Kelly in 1942, has attracted much research attention and yet remains one of the foremost unsolved problems in graph theory. Recently, Bastide, Cook, Erickson, Groenland, Kreveld, Mannens, and Vermeulen proposed a new model of partial informati…
▽ More
The problem of graph reconstruction has been studied in its various forms over the years. In particular, the Reconstruction Conjecture, proposed by Ulam and Kelly in 1942, has attracted much research attention and yet remains one of the foremost unsolved problems in graph theory. Recently, Bastide, Cook, Erickson, Groenland, Kreveld, Mannens, and Vermeulen proposed a new model of partial information, where we are given the set of connected triples T_3, which is the set of 3-subsets of the vertex set that induce connected subgraphs. They proved that reconstruction is unique within the class of triangle-free graphs, 2-connected outerplanar graphs, and maximal planar graphs. They also showed that almost every graph can be uniquely reconstructed from their connected triples. However, little is known about other classes of non-triangle-free graphs within which reconstruction can occur uniquely, nor do we understand what kind of graphs can be uniquely reconstructed from their connected triples without assuming anything about the classes of graphs to which they belong.
The main result of this paper is a complete characterization of all graphs that can be uniquely reconstructed from their connected triples T_3. We also show that reconstruction from T_3 is unique within the class of regular planar graphs, 5-connected planar graphs, certain strongly regular graphs, and complete multi-partite graphs, whereas it is not unique for the class of k-connected planar graphs with k less or equal to 4, Eulerian graphs, or Hamiltonian graphs.
△ Less
Submitted 18 September, 2023;
originally announced September 2023.
-
Symmetries of equivariant Khovanov-Rozansky homology
Authors:
You Qi,
Louis-Hadrien Robert,
Joshua Sussan,
Emmanuel Wagner
Abstract:
We construct an $\mathfrak{sl}_2$-action on equivariant $\mathfrak{gl}_N$-link homologies. As a consequence we obtain an action of $\mathfrak{sl}_2$ on these homologies as well as a $p$-DG structures for $p$ a prime number. We explore topological applications of these structures.
We construct an $\mathfrak{sl}_2$-action on equivariant $\mathfrak{gl}_N$-link homologies. As a consequence we obtain an action of $\mathfrak{sl}_2$ on these homologies as well as a $p$-DG structures for $p$ a prime number. We explore topological applications of these structures.
△ Less
Submitted 19 July, 2023; v1 submitted 19 June, 2023;
originally announced June 2023.
-
Probabilistic computation and uncertainty quantification with emerging covariance
Authors:
Hengyuan Ma,
Yang Qi,
Li Zhang,
Wenlian Lu,
Jianfeng Feng
Abstract:
Building robust, interpretable, and secure AI system requires quantifying and representing uncertainty under a probabilistic perspective to mimic human cognitive abilities. However, probabilistic computation presents significant challenges for most conventional artificial neural network, as they are essentially implemented in a deterministic manner. In this paper, we develop an efficient probabili…
▽ More
Building robust, interpretable, and secure AI system requires quantifying and representing uncertainty under a probabilistic perspective to mimic human cognitive abilities. However, probabilistic computation presents significant challenges for most conventional artificial neural network, as they are essentially implemented in a deterministic manner. In this paper, we develop an efficient probabilistic computation framework by truncating the probabilistic representation of neural activation up to its mean and covariance and construct a moment neural network that encapsulates the nonlinear coupling between the mean and covariance of the underlying stochastic network. We reveal that when only the mean but not the covariance is supervised during gradient-based learning, the unsupervised covariance spontaneously emerges from its nonlinear coupling with the mean and faithfully captures the uncertainty associated with model predictions. Our findings highlight the inherent simplicity of probabilistic computation by seamlessly incorporating uncertainty into model prediction, paving the way for integrating it into large-scale AI systems.
△ Less
Submitted 12 January, 2024; v1 submitted 30 May, 2023;
originally announced May 2023.
-
A braid group action on an $A_\infty$-category for zigzag algebras
Authors:
Benjamin Cooper,
You Qi,
Joshua Sussan
Abstract:
We construct differential graded enhancements of the zigzag algebras which were used by Khovanov, Seidel and Thomas to produce categorical braid group actions. These enhancements are related to $p$-differential graded structures by a version of Koszul duality. We prove that the minimal model $A_\infty$-structure on the zigzag algebras is {\em not} formal. We construct a braid group action in this…
▽ More
We construct differential graded enhancements of the zigzag algebras which were used by Khovanov, Seidel and Thomas to produce categorical braid group actions. These enhancements are related to $p$-differential graded structures by a version of Koszul duality. We prove that the minimal model $A_\infty$-structure on the zigzag algebras is {\em not} formal. We construct a braid group action in this setting and suggest a symplectic interpretation.
△ Less
Submitted 4 May, 2023;
originally announced May 2023.
-
Feature Affinity Assisted Knowledge Distillation and Quantization of Deep Neural Networks on Label-Free Data
Authors:
Zhijian Li,
Biao Yang,
Penghang Yin,
Yingyong Qi,
Jack Xin
Abstract:
In this paper, we propose a feature affinity (FA) assisted knowledge distillation (KD) method to improve quantization-aware training of deep neural networks (DNN). The FA loss on intermediate feature maps of DNNs plays the role of teaching middle steps of a solution to a student instead of only giving final answers in the conventional KD where the loss acts on the network logits at the output leve…
▽ More
In this paper, we propose a feature affinity (FA) assisted knowledge distillation (KD) method to improve quantization-aware training of deep neural networks (DNN). The FA loss on intermediate feature maps of DNNs plays the role of teaching middle steps of a solution to a student instead of only giving final answers in the conventional KD where the loss acts on the network logits at the output level. Combining logit loss and FA loss, we found that the quantized student network receives stronger supervision than from the labeled ground-truth data. The resulting FAQD is capable of compressing model on label-free data, which brings immediate practical benefits as pre-trained teacher models are readily available and unlabeled data are abundant. In contrast, data labeling is often laborious and expensive. Finally, we propose a fast feature affinity (FFA) loss that accurately approximates FA loss with a lower order of computational complexity, which helps speed up training for high resolution image input.
△ Less
Submitted 18 August, 2023; v1 submitted 9 February, 2023;
originally announced February 2023.
-
Symmetries of $\mathfrak{gl}_N$-foams
Authors:
You Qi,
Louis-Hadrien Robert,
Joshua Sussan,
Emmanuel Wagner
Abstract:
We give an action of a Lie subalgebra of the Witt algebra on foams. This action is compatible with the $\mathfrak{gl}_N$-foam evaluation formula. In particular, this endows states spaces associated with $\mathfrak{gl}_N$-webs with an $\mathfrak{sl}_2$-action. When working in positive characteristic, this can be used to define a $p$-DG structure on these state spaces.
We give an action of a Lie subalgebra of the Witt algebra on foams. This action is compatible with the $\mathfrak{gl}_N$-foam evaluation formula. In particular, this endows states spaces associated with $\mathfrak{gl}_N$-webs with an $\mathfrak{sl}_2$-action. When working in positive characteristic, this can be used to define a $p$-DG structure on these state spaces.
△ Less
Submitted 16 January, 2024; v1 submitted 20 December, 2022;
originally announced December 2022.
-
An efficient numerical algorithm for the moment neural activation
Authors:
Yang Qi
Abstract:
Derived from spiking neuron models via the diffusion approximation, the moment activation (MA) faithfully captures the nonlinear coupling of correlated neural variability. However, numerical evaluation of the MA faces significant challenges due to a number of ill-conditioned Dawson-like functions. By deriving asymptotic expansions of these functions, we develop an efficient numerical algorithm for…
▽ More
Derived from spiking neuron models via the diffusion approximation, the moment activation (MA) faithfully captures the nonlinear coupling of correlated neural variability. However, numerical evaluation of the MA faces significant challenges due to a number of ill-conditioned Dawson-like functions. By deriving asymptotic expansions of these functions, we develop an efficient numerical algorithm for evaluating the MA and its derivatives ensuring reliability, speed, and accuracy. We also provide exact analytical expressions for the MA in the weak fluctuation limit. Powered by this efficient algorithm, the MA may serve as an effective tool for investigating the dynamics of correlated neural variability in large-scale spiking neural circuits.
△ Less
Submitted 2 December, 2022;
originally announced December 2022.
-
Limiting distributions of the likelihood ratio test statistics for independence of normal random vectors
Authors:
Mingyue Hu,
Yongcheng Qi
Abstract:
Consider the likelihood ratio test (LRT) statistics for the independence of sub-vectors from a $p$-variate normal random vector. We are devoted to deriving the limiting distributions of the LRT statistics based on a random sample of size $n$. It is well known that the limit is chi-square distribution when the dimension of the data or the number of the parameters are fixed. In a recent work by Qi,…
▽ More
Consider the likelihood ratio test (LRT) statistics for the independence of sub-vectors from a $p$-variate normal random vector. We are devoted to deriving the limiting distributions of the LRT statistics based on a random sample of size $n$. It is well known that the limit is chi-square distribution when the dimension of the data or the number of the parameters are fixed. In a recent work by Qi, Wang and Zhang (Ann Inst Stat Math (2019) 71: 911--946), it was shown that the LRT statistics are asymptotically normal under condition that the lengths of the normal random sub-vectors are relatively balanced if the dimension $p$ goes to infinity with the sample size $n$. In this paper, we investigate the limiting distributions of the LRT statistic under general conditions. We find out all types of limiting distributions and obtain the necessary and sufficient conditions for the LRT statistic to converge to a normal distribution when $p$ goes to infinity. We also investigate the limiting distribution of the adjusted LRT test statistic proposed in Qi, Wang and Zhang (2019). Moreover, we present simulation results to compare the performance of classical chi-square approximation, normal and non-normal approximation to the LRT statistics, chi-square approximation to the adjusted test statistic, and some other test statistics.
△ Less
Submitted 20 July, 2022;
originally announced July 2022.
-
Inference of high quantiles of a heavy-tailed distribution from block data
Authors:
Yongcheng Qi,
Mengzi Xie,
**g** Yang
Abstract:
In this paper we consider the estimation problem for high quantiles of a heavy-tailed distribution from block data when only a few largest values are observed within blocks. We propose estimators for high quantiles and prove that these estimators are asymptotically normal. Furthermore, we employ empirical likelihood method and adjusted empirical likelihood method to constructing the confidence int…
▽ More
In this paper we consider the estimation problem for high quantiles of a heavy-tailed distribution from block data when only a few largest values are observed within blocks. We propose estimators for high quantiles and prove that these estimators are asymptotically normal. Furthermore, we employ empirical likelihood method and adjusted empirical likelihood method to constructing the confidence intervals of high quantiles. Through a simulation study we also compare the performance of the normal approximation method and the adjusted empirical likelihood methods in terms of the coverage probability and length of the confidence intervals.
△ Less
Submitted 24 June, 2023; v1 submitted 16 July, 2022;
originally announced July 2022.
-
A new type of results on probabilities of moderate deviations for i.i.d. random variables
Authors:
Deli Li,
Yu Miao,
Yongcheng Qi
Abstract:
Let $\{X, X_{n}; n \geq 1\}$ be a sequence of i.i.d. non-degenerate real-valued random variables with $\mathbb{E}X^{2} < \infty$. Let $S_{n} = \sum_{i=1}^{n} X_{i}$, $n \geq 1$. Let $g(\cdot): ~[0, \infty) \rightarrow [0, \infty)$ be a nondecreasing regularly varying function with index $ρ\geq 0$ and $\lim_{t \rightarrow \infty} g(t) = \infty$. Let $μ= \mathbb{E}X$ and…
▽ More
Let $\{X, X_{n}; n \geq 1\}$ be a sequence of i.i.d. non-degenerate real-valued random variables with $\mathbb{E}X^{2} < \infty$. Let $S_{n} = \sum_{i=1}^{n} X_{i}$, $n \geq 1$. Let $g(\cdot): ~[0, \infty) \rightarrow [0, \infty)$ be a nondecreasing regularly varying function with index $ρ\geq 0$ and $\lim_{t \rightarrow \infty} g(t) = \infty$. Let $μ= \mathbb{E}X$ and $σ^{2} = \mathbb{E}(X - μ)^{2}$. In this paper, we obtain precise asymptotic estimates for probabilities of moderate deviations by showing that, for all $x > 0$, \[ \limsup_{n \rightarrow \infty} \frac{\log \mathbb{P}\left(S_{n} - n μ> x \sqrt{n g(\log n)} \right)}{g(\log n)} = - \left(\frac{x^{2}}{2σ^{2}} \wedge \frac{\overlineλ_{1}}{2^ρ} \right), \] \[ \liminf_{n \rightarrow \infty} \frac{\log \mathbb{P}\left(S_{n} - n μ> x \sqrt{n g(\log n)} \right)}{g(\log n)} = - \left(\frac{x^{2}}{2σ^{2}} \wedge \frac{\underlineλ_{1}}{2^ρ} \right), \] \[ \limsup_{n \rightarrow \infty} \frac{\log \mathbb{P}\left(S_{n} - n μ< -x \sqrt{n g(\log n)} \right)}{g(\log n)} = - \left(\frac{x^{2}}{2σ^{2}} \wedge \frac{\overlineλ_{2}}{2^ρ} \right), \] and \[ \liminf_{n \rightarrow \infty} \frac{\log \mathbb{P}\left(S_{n} - n μ< -x \sqrt{n g(\log n)} \right)}{g(\log n)} = - \left(\frac{x^{2}}{2σ^{2}} \wedge \frac{\underlineλ_{2}}{2^ρ} \right), \] where $\overlineλ_{1}$ are $\underlineλ_{1}$ are determined by the asymptotic behavior of $\mathbb{P}(X > t)$ and $\overlineλ_{2}$ and $\underlineλ_{2}$ are determined by the asymptotic behavior of $\mathbb{P}(X < -t)$. Unlike those known results in the literature, the moderate deviation results established in this paper depend on both the variance and the asymptotic behavior of the tail distribution of $X$.
△ Less
Submitted 7 July, 2022;
originally announced July 2022.
-
A Rickard equivalence for hopfological homotopy categories
Authors:
You Qi
Abstract:
In his paper [Ric89], Rickard presents the stable module category of a self-injective algebra as a Verdier quotient of its derived category by perfect complexes. We present a similar realization of the homotopy category in hopfological algebra as such a Verdier quotient.
In his paper [Ric89], Rickard presents the stable module category of a self-injective algebra as a Verdier quotient of its derived category by perfect complexes. We present a similar realization of the homotopy category in hopfological algebra as such a Verdier quotient.
△ Less
Submitted 17 November, 2022; v1 submitted 29 April, 2022;
originally announced April 2022.
-
Empirical likelihood method for complete independence test on high dimensional data
Authors:
Yongcheng Qi,
Yingchao Zhou
Abstract:
Given a random sample of size $n$ from a $p$ dimensional random vector, where both $n$ and $p$ are large, we are interested in testing whether the $p$ components of the random vector are mutually independent. This is the so-called complete independence test. In the multivariate normal case, it is equivalent to testing whether the correlation matrix is an identity matrix. In this paper, we propose…
▽ More
Given a random sample of size $n$ from a $p$ dimensional random vector, where both $n$ and $p$ are large, we are interested in testing whether the $p$ components of the random vector are mutually independent. This is the so-called complete independence test. In the multivariate normal case, it is equivalent to testing whether the correlation matrix is an identity matrix. In this paper, we propose a one-sided empirical likelihood method for the complete independence test for multivariate normal data based on squared sample correlation coefficients. The limiting distribution for our one-sided empirical likelihood test statistic is proved to be $Z^2I(Z>0)$ when both $n$ and $p$ tend to infinity, where $Z$ is a standard normal random variable. In order to improve the power of the empirical likelihood test statistic, we also introduce a rescaled empirical likelihood test statistic. We carry out an extensive simulation study to compare the performance of the rescaled empirical likelihood method and two other statistics which are related to the sum of squared sample correlation coefficients.
△ Less
Submitted 20 January, 2022;
originally announced January 2022.
-
A categorification of the colored Jones polynomial at a root of unity
Authors:
You Qi,
Louis-Hadrien Robert,
Joshua Sussan,
Emmanuel Wagner
Abstract:
There is a $p$-differential on the triply-graded Khovanov--Rozansky homology of knots and links over a field of positive characteristic $p$ that gives rise to an invariant in the homotopy category finite-dimensional $p$-complexes.
A differential on triply-graded homology discovered by Cautis is compatible with the $p$-differential structure. As a consequence we get a categorification of the colo…
▽ More
There is a $p$-differential on the triply-graded Khovanov--Rozansky homology of knots and links over a field of positive characteristic $p$ that gives rise to an invariant in the homotopy category finite-dimensional $p$-complexes.
A differential on triply-graded homology discovered by Cautis is compatible with the $p$-differential structure. As a consequence we get a categorification of the colored Jones polynomial evaluated at a $2p$th root of unity.
△ Less
Submitted 25 November, 2021;
originally announced November 2021.
-
Asymptotic Distributions for Likelihood Ratio Tests for the Equality of Covariance Matrices
Authors:
Wenchuan Guo,
Yongcheng Qi
Abstract:
Consider $k$ independent random samples from $p$-dimensional multivariate normal distributions. We are interested in the limiting distribution of the log-likelihood ratio test statistics for testing for the equality of $k$ covariance matrices. It is well known from classical multivariate statistics that the limit is a chi-square distribution when $k$ and $p$ are fixed integers. Jiang and Yang~\cit…
▽ More
Consider $k$ independent random samples from $p$-dimensional multivariate normal distributions. We are interested in the limiting distribution of the log-likelihood ratio test statistics for testing for the equality of $k$ covariance matrices. It is well known from classical multivariate statistics that the limit is a chi-square distribution when $k$ and $p$ are fixed integers. Jiang and Yang~\cite{JY13} and Jiang and Qi~\cite{JQ15} have obtained the central limit theorem for the log-likelihood ratio test statistics when the dimensionality $p$ goes to infinity with the sample sizes. In this paper, we derive the central limit theorem when either $p$ or $k$ goes to infinity. We also propose adjusted test statistics which can be well approximated by chi-squared distributions regardless of values for $p$ and $k$. Furthermore, we present numerical simulation results to evaluate the performance of our adjusted test statistics and the log-likelihood ratio statistics based on classical chi-square approximation and the normal approximation.
△ Less
Submitted 19 May, 2023; v1 submitted 5 October, 2021;
originally announced October 2021.
-
On some $p$-differential graded link homologies II
Authors:
You Qi,
Joshua Sussan
Abstract:
In arXiv:2009.06498, a link invariant categorifying the Jones polynomial at a $2p$th root of unity, where $p$ is an odd prime, was constructed. This categorification utilized an $N=2$ specialization of a differential introduced by Cautis. Here we give a family of link homologies where the Cautis differential is specialized to a positive integer of the form $N=kp+2$. When $k$ is even, all these lin…
▽ More
In arXiv:2009.06498, a link invariant categorifying the Jones polynomial at a $2p$th root of unity, where $p$ is an odd prime, was constructed. This categorification utilized an $N=2$ specialization of a differential introduced by Cautis. Here we give a family of link homologies where the Cautis differential is specialized to a positive integer of the form $N=kp+2$. When $k$ is even, all these link homologies categorify the Jones polynomial evaluated at a $2p$th root of unity, but they are non-isomorphic invariants.
△ Less
Submitted 24 August, 2021;
originally announced August 2021.
-
Tropical linear regression and mean payoff games: or, how to measure the distance to equilibria
Authors:
Marianne Akian,
Stéphane Gaubert,
Yang Qi,
Omar Saadi
Abstract:
We study a tropical linear regression problem consisting in finding the best approximation of a set of points by a tropical hyperplane. We establish a strong duality theorem, showing that the value of this problem coincides with the maximal radius of a Hilbert's ball included in a tropical polyhedron. We also show that this regression problem is polynomial-time equivalent to mean payoff games. We…
▽ More
We study a tropical linear regression problem consisting in finding the best approximation of a set of points by a tropical hyperplane. We establish a strong duality theorem, showing that the value of this problem coincides with the maximal radius of a Hilbert's ball included in a tropical polyhedron. We also show that this regression problem is polynomial-time equivalent to mean payoff games. We illustrate our results by solving an inverse problem from auction theory. In this setting, a tropical hyperplane represents the set of equilibrium prices. Tropical linear regression allows us to quantify the distance of a market to the set of equilibria, and infer secret preferences of a decision maker.
△ Less
Submitted 21 June, 2021; v1 submitted 3 June, 2021;
originally announced June 2021.
-
Limiting Empirical Spectral Distribution for Products of Rectangular Matrices
Authors:
Yongcheng Qi,
Hongru Zhao
Abstract:
In this paper, we consider $m$ independent random rectangular matrices whose entries are independent and identically distributed standard complex Gaussian random variables and assume the product of the $m$ rectangular matrices is an $n$ by $n$ square matrix. We study the limiting empirical spectral distributions of the product where the dimension of the product matrix goes to infinity, and $m$ may…
▽ More
In this paper, we consider $m$ independent random rectangular matrices whose entries are independent and identically distributed standard complex Gaussian random variables and assume the product of the $m$ rectangular matrices is an $n$ by $n$ square matrix. We study the limiting empirical spectral distributions of the product where the dimension of the product matrix goes to infinity, and $m$ may change with the dimension of the product matrix and diverge. We give a complete description for the limiting distribution of the empirical spectral distributions for the product matrix and illustrate some examples.
△ Less
Submitted 7 April, 2021;
originally announced April 2021.
-
Actions of sl_2 on algebras appearing in categorification
Authors:
Ben Elias,
You Qi
Abstract:
We prove that many of the recently-constructed algebras and categories which appear in categorification can be equipped with an action of $\mathfrak{sl}_2$ by derivations. The $\mathfrak{sl}_2$ representations which appear are filtered by tensor products of coverma modules. In a future paper, we will address the implications of the $\mathfrak{sl}_2$ structure for categorification.
We prove that many of the recently-constructed algebras and categories which appear in categorification can be equipped with an action of $\mathfrak{sl}_2$ by derivations. The $\mathfrak{sl}_2$ representations which appear are filtered by tensor products of coverma modules. In a future paper, we will address the implications of the $\mathfrak{sl}_2$ structure for categorification.
△ Less
Submitted 22 April, 2022; v1 submitted 26 February, 2021;
originally announced March 2021.
-
A braid group action on a $p$-DG homotopy category
Authors:
You Qi,
Joshua Sussan,
Yasuyoshi Yonezawa
Abstract:
We construct a braid group action on a homotopy category of $p$-DG modules of a deformed Webster algebra.
We construct a braid group action on a homotopy category of $p$-DG modules of a deformed Webster algebra.
△ Less
Submitted 9 February, 2022; v1 submitted 30 December, 2020;
originally announced December 2020.
-
Numerical Calabi-Yau metrics from holomorphic networks
Authors:
Michael R. Douglas,
Subramanian Lakshminarasimhan,
Yidi Qi
Abstract:
We propose machine learning inspired methods for computing numerical Calabi-Yau (Ricci flat Kähler) metrics, and implement them using Tensorflow/Keras. We compare them with previous work, and find that they are far more accurate for manifolds with little or no symmetry. We also discuss issues such as overparameterization and choice of optimization methods.
We propose machine learning inspired methods for computing numerical Calabi-Yau (Ricci flat Kähler) metrics, and implement them using Tensorflow/Keras. We compare them with previous work, and find that they are far more accurate for manifolds with little or no symmetry. We also discuss issues such as overparameterization and choice of optimization methods.
△ Less
Submitted 5 May, 2021; v1 submitted 8 December, 2020;
originally announced December 2020.
-
On some $p$-differential graded link homologies
Authors:
You Qi,
Joshua Sussan
Abstract:
We show that the triply graded Khovanov-Rozansky homology of knots and links over a field of positive odd characteristic $p$ descends to an invariant in the homotopy category finite-dimensional $p$-complexes.
A $p$-extended differential on the triply graded homology discovered by Cautis is compatible with the $p$-DG structure. As a consequence we get a categorification of the Jones polynomial ev…
▽ More
We show that the triply graded Khovanov-Rozansky homology of knots and links over a field of positive odd characteristic $p$ descends to an invariant in the homotopy category finite-dimensional $p$-complexes.
A $p$-extended differential on the triply graded homology discovered by Cautis is compatible with the $p$-DG structure. As a consequence we get a categorification of the Jones polynomial evaluated at an odd prime root of unity
△ Less
Submitted 19 December, 2022; v1 submitted 14 September, 2020;
originally announced September 2020.
-
Evaluating thin flat surfaces
Authors:
Mikhail Khovanov,
You Qi,
Lev Rozansky
Abstract:
We consider recognizable evaluations for a suitable category of oriented two-dimensional cobordisms with corners between finite unions of intervals. We call such cobordisms thin flat surfaces. An evaluation is given by a power series in two variables. Recognizable evaluations correspond to series that are ratios of a two-variable polynomial by the product of two one-variable polynomials, one for e…
▽ More
We consider recognizable evaluations for a suitable category of oriented two-dimensional cobordisms with corners between finite unions of intervals. We call such cobordisms thin flat surfaces. An evaluation is given by a power series in two variables. Recognizable evaluations correspond to series that are ratios of a two-variable polynomial by the product of two one-variable polynomials, one for each variable. They are also in a bijection with isomorphism classes of commutative Frobenius algebras on two generators with a nondegenerate trace fixed. The latter algebras of dimension n correspond to points on the dual tautological bundle on the Hilbert scheme of n points on the affine plane, with a certain divisor removed from the bundle. A recognizable evaluation gives rise to a functor from the above cobordism category of thin flat surfaces to the category of finite-dimensional vector spaces. These functors may be non-monoidal in interesting cases. To a recognizable evaluation we also assign an analogue of the Deligne category and of its quotient by the ideal of negligible morphisms.
△ Less
Submitted 2 September, 2020;
originally announced September 2020.
-
Limiting Spectral Radii of Circular Unitary Matrices under Light Truncation
Authors:
Yu Miao,
Yongcheng Qi
Abstract:
Consider a truncated circular unitary matrix which is a $p_n$ by $p_n$ submatrix of an $n$ by $n$ circular unitary matrix after deleting the last $n-p_n$ columns and rows. Jiang and Qi \cite{JiangQi2017} and Gui and Qi \cite{GQ2018} study the limiting distributions of the maximum absolute value of the eigenvalues (known as spectral radius) of the truncated matrix. Some limiting distributions for t…
▽ More
Consider a truncated circular unitary matrix which is a $p_n$ by $p_n$ submatrix of an $n$ by $n$ circular unitary matrix after deleting the last $n-p_n$ columns and rows. Jiang and Qi \cite{JiangQi2017} and Gui and Qi \cite{GQ2018} study the limiting distributions of the maximum absolute value of the eigenvalues (known as spectral radius) of the truncated matrix. Some limiting distributions for the spectral radius for the truncated circular unitary matrix have been obtained under the following conditions: (1). $p_n/n$ is bounded away from $0$ and $1$; (2). $p_n\to\infty$ and $p_n/n\to 0$ as $n\to\infty$; (3). $(n-p_n)/n\to 0$ and $(n-p_n)/(\log n)^3\to\infty$ as $n\to\infty$; (4). $n-p_n\to\infty$ and $(n-p_n)/\log n\to 0$ as $n\to\infty$; and (5). $n-p_n=k\ge 1$ is a fixed integer. The spectral radius converges in distribution to the Gumbel distribution under the first four conditions and to a reversed Weibull distribution under the fifth condition. Apparently, the conditions above do not cover the case when $n-p_n$ is of order between $\log n$ and $(\log n)^3$. In this paper, we prove that the spectral radius converges in distribution to the Gumbel distribution as well in this case, as conjectured by Gui and Qi \cite{GQ2018}.
△ Less
Submitted 23 August, 2020;
originally announced August 2020.
-
Categorifying Hecke algebras at prime roots of unity, part I
Authors:
Ben Elias,
You Qi
Abstract:
We equip the type $A$ diagrammatic Hecke category with a special derivation, so that after specialization to characteristic $p$ it becomes a $p$-dg category. We prove that the defining relations of the Hecke algebra are satisfied in the $p$-dg Grothendieck group. We conjecture that the $p$-dg Grothendieck group is isomorphic to the Iwahori-Hecke algebra, equip** it with a basis which may differ…
▽ More
We equip the type $A$ diagrammatic Hecke category with a special derivation, so that after specialization to characteristic $p$ it becomes a $p$-dg category. We prove that the defining relations of the Hecke algebra are satisfied in the $p$-dg Grothendieck group. We conjecture that the $p$-dg Grothendieck group is isomorphic to the Iwahori-Hecke algebra, equip** it with a basis which may differ from both the Kazhdan-Lusztig basis and the $p$-canonical basis. More precise conjectures will be found in the sequel.
Here are some other results contained in this paper. We provide an incomplete proof of the classification of all degree $+2$ derivations on the diagrammatic Hecke category, and a complete proof of the classification of those derivations for which the defining relations of the Hecke algebra are satisfied in the $p$-dg Grothendieck group. In particular, our special derivation is unique up to duality and equivalence. We prove that no such derivation exists in simply-laced types outside of finite and affine type $A$. We also examine a particular Bott-Samelson bimodule in type $A_7$, which is indecomposable in characteristic $2$ but decomposable in all other characteristics. We prove that this Bott-Samelson bimodule admits no nontrivial fantastic filtrations in any characteristic, which is the analogue in the $p$-dg setting of being indecomposable.
△ Less
Submitted 21 February, 2023; v1 submitted 6 May, 2020;
originally announced May 2020.
-
Universal commensurability augmented Teichmüller space and moduli space
Authors:
Guangming Hu,
Hideki Miyachi,
Yi Qi
Abstract:
It is known that every finitely unbranched covering $α:\widetilde{S}_{g(α)}\rightarrow S$ of a compact Riemann surface $S$ with genus $g\geq2$ induces an isometric embedding $Γ_α$ from the Teichmüller space $T(S)$ to the Teichüller space $T(\widetilde{S}_{g(α)})$. Actually, it has been showed that the isometric embedding $Γ_α$ can be extended isometrically to the augmented Teichmüller space…
▽ More
It is known that every finitely unbranched covering $α:\widetilde{S}_{g(α)}\rightarrow S$ of a compact Riemann surface $S$ with genus $g\geq2$ induces an isometric embedding $Γ_α$ from the Teichmüller space $T(S)$ to the Teichüller space $T(\widetilde{S}_{g(α)})$. Actually, it has been showed that the isometric embedding $Γ_α$ can be extended isometrically to the augmented Teichmüller space $\widehat{T}(S)$ of $T(S)$. Using this result, we construct a directed limit $\widehat{T}_{\infty}(S)$ of augmented Teichmüller spaces, where the index runs over all finitely unbranched coverings of $S$. Then, we show that the action of the universal commensurability modular group $Mod_{\infty}(S)$ can extend isometrically on $\widehat{T}_{\infty}(S)$. Furthermore, for any $X_{\infty}\in T_{\infty}(S)$, its orbit of the action of the universal commensurability modular group $Mod_{\infty}(S)$ on the universal commensurability augmented Teichmüller space $\widehat{T}_{\infty}(S)$ is dense. Finally, we also construct a directed limit $\widehat{M}_{\infty}(S)$ of augmented moduli spaces by characteristic towers and show that the subgroup $Caut(π_{1}(S))$ of $Mod_{\infty}(S)$ acts on $\widehat{T}_{\infty}(S)$ to produce $\widehat{M}_{\infty}(S)$ as the quotient.
△ Less
Submitted 14 April, 2020; v1 submitted 5 April, 2020;
originally announced April 2020.
-
Remarks on the derived center of small quantum groups
Authors:
Anna Lachowska,
You Qi
Abstract:
Let $\mathsf{u}_q(\mathfrak{g})$ be the small quantum group associated with a complex semisimple Lie algebra $\mathfrak{g}$ and a primitive root of unity q, satisfying certain restrictions. We establish the equivalence between three different actions of $\mathfrak{g}$ on the center of $\mathsf{u}_q(\mathfrak{g})$ and on the higher derived center of $\mathsf{u}_q(\mathfrak{g})$. Based on the trivia…
▽ More
Let $\mathsf{u}_q(\mathfrak{g})$ be the small quantum group associated with a complex semisimple Lie algebra $\mathfrak{g}$ and a primitive root of unity q, satisfying certain restrictions. We establish the equivalence between three different actions of $\mathfrak{g}$ on the center of $\mathsf{u}_q(\mathfrak{g})$ and on the higher derived center of $\mathsf{u}_q(\mathfrak{g})$. Based on the triviality of this action for $\mathfrak{g} = \mathfrak{sl}_2, \mathfrak{sl}_3, \mathfrak{sl}_4$, we conjecture that, in finite type A, central elements of the small quantum group $\mathsf{u}_q(\mathfrak{sl}_n)$ arise as the restriction of central elements in the big quantum group $\mathsf{U}_q(\mathfrak{sl}_n)$.
We also study the role of an ideal $\mathsf{z}_{\mathrm{Hig}}$ known as the Higman ideal in the center of $\mathsf{u}_q(\mathfrak{g})$. We show that it coincides with the intersection of the Harish-Chandra center and its Fourier transform, and compute the dimension of $\mathsf{z}_{\mathrm{Hig}}$ in type A. As an illustration we provide a detailed explicit description of the derived center of $\mathsf{u}_q(\mathfrak{sl}_2)$ and its various symmetries.
△ Less
Submitted 10 July, 2021; v1 submitted 18 December, 2019;
originally announced December 2019.
-
A faithful braid group action on the stable category of tricomplexes
Authors:
Mikhail Khovanov,
You Qi
Abstract:
Bicomplexes of vector spaces frequently appear throughout algebra and geometry. In Section 2 we explain how to think about the arrows in the spectral sequence of a bicomplex via its indecomposable summands. Polycomplexes seem to be much more rare. In Section 3 of this paper we rethink a well-known faithful categorical braid group action via an action on the stable category of tricomplexes.
Bicomplexes of vector spaces frequently appear throughout algebra and geometry. In Section 2 we explain how to think about the arrows in the spectral sequence of a bicomplex via its indecomposable summands. Polycomplexes seem to be much more rare. In Section 3 of this paper we rethink a well-known faithful categorical braid group action via an action on the stable category of tricomplexes.
△ Less
Submitted 30 March, 2020; v1 submitted 6 November, 2019;
originally announced November 2019.
-
Spectral Radii of Products of Random Rectangular Matrices
Authors:
Yongcheng Qi,
Mengzi Xie
Abstract:
We consider m independent random rectangular matrices whose entries are independent and identically distributed standard complex Gaussian random variables. Assume the product of the m rectangular matrices is an n by n square matrix. The maximum absolute values of the n eigenvalues of the product matrix is called spectral radius. In this paper, we study the limiting spectral radii of the product wh…
▽ More
We consider m independent random rectangular matrices whose entries are independent and identically distributed standard complex Gaussian random variables. Assume the product of the m rectangular matrices is an n by n square matrix. The maximum absolute values of the n eigenvalues of the product matrix is called spectral radius. In this paper, we study the limiting spectral radii of the product when m changes with n and can even diverge. We give a complete description for the limiting distribution of the spectral radius. Our results reduce to those in Jiang and Qi [26] when the rectangular matrices are square ones.
△ Less
Submitted 15 July, 2022; v1 submitted 11 September, 2019;
originally announced September 2019.
-
Understanding Straight-Through Estimator in Training Activation Quantized Neural Nets
Authors:
Penghang Yin,
Jiancheng Lyu,
Shuai Zhang,
Stanley Osher,
Yingyong Qi,
Jack Xin
Abstract:
Training activation quantized neural networks involves minimizing a piecewise constant function whose gradient vanishes almost everywhere, which is undesirable for the standard back-propagation or chain rule. An empirical way around this issue is to use a straight-through estimator (STE) (Bengio et al., 2013) in the backward pass only, so that the "gradient" through the modified chain rule becomes…
▽ More
Training activation quantized neural networks involves minimizing a piecewise constant function whose gradient vanishes almost everywhere, which is undesirable for the standard back-propagation or chain rule. An empirical way around this issue is to use a straight-through estimator (STE) (Bengio et al., 2013) in the backward pass only, so that the "gradient" through the modified chain rule becomes non-trivial. Since this unusual "gradient" is certainly not the gradient of loss function, the following question arises: why searching in its negative direction minimizes the training loss? In this paper, we provide the theoretical justification of the concept of STE by answering this question. We consider the problem of learning a two-linear-layer network with binarized ReLU activation and Gaussian input data. We shall refer to the unusual "gradient" given by the STE-modifed chain rule as coarse gradient. The choice of STE is not unique. We prove that if the STE is properly chosen, the expected coarse gradient correlates positively with the population gradient (not available for the training), and its negation is a descent direction for minimizing the population loss. We further show the associated coarse gradient descent algorithm converges to a critical point of the population loss minimization problem. Moreover, we show that a poor choice of STE leads to instability of the training algorithm near certain local minima, which is verified with CIFAR-10 experiments.
△ Less
Submitted 25 September, 2019; v1 submitted 13 March, 2019;
originally announced March 2019.
-
p-DG cyclotomic nilHecke algebras II
Authors:
You Qi,
Joshua Sussan
Abstract:
We categorify tensor products of the fundamental representation of quantum $\mathfrak{sl}_2$ at prime roots of unity building upon earlier work where a tensor product of two Weyl modules was categorified.
We categorify tensor products of the fundamental representation of quantum $\mathfrak{sl}_2$ at prime roots of unity building upon earlier work where a tensor product of two Weyl modules was categorified.
△ Less
Submitted 1 December, 2021; v1 submitted 11 November, 2018;
originally announced November 2018.
-
Topology of tensor ranks
Authors:
Pierre Comon,
Lek-Heng Lim,
Yang Qi,
Ke Ye
Abstract:
We study path-connectedness and homotopy groups of sets of tensors defined by tensor rank, border rank, multilinear rank, as well as their symmetric counterparts for symmetric tensors. We show that over $\mathbb{C}$, the set of rank-$r$ tensors and the set of symmetric rank-$r$ symmetric tensors are both path-connected if $r$ is not more than the complex generic rank; these results also extend to…
▽ More
We study path-connectedness and homotopy groups of sets of tensors defined by tensor rank, border rank, multilinear rank, as well as their symmetric counterparts for symmetric tensors. We show that over $\mathbb{C}$, the set of rank-$r$ tensors and the set of symmetric rank-$r$ symmetric tensors are both path-connected if $r$ is not more than the complex generic rank; these results also extend to border rank and symmetric border rank over $\mathbb{C}$. Over $\mathbb{R}$, the set of rank-$r$ tensors is path-connected if it has the expected dimension but the corresponding result for symmetric rank-$r$ symmetric $d$-tensors depends on the order $d$: connected when $d$ is odd but not when $d$ is even. Border rank and symmetric border rank over $\mathbb{R}$ have essentially the same path-connectedness properties as rank and symmetric rank over $\mathbb{R}$. When $r$ is greater than the complex generic rank, we are unable to discern any general pattern: For example, we show that border-rank-three tensors in $\mathbb{R}^2 \otimes \mathbb{R}^2 \otimes \mathbb{R}^2$ fall into four connected components. For multilinear rank, the manifold of $d$-tensors of multilinear rank $(r_1,\dots,r_d)$ in $\mathbb{C}^{n_1} \otimes \cdots \otimes \mathbb{C}^{n_d}$ is always path-connected, and the same is true in $\mathbb{R}^{n_1} \otimes \cdots \otimes \mathbb{R}^{n_d}$ unless $n_i = r_i = \prod_{j \ne i} r_j$ for some $i\in\{1, \dots, d\}$. Beyond path-connectedness, we determine, over both $\mathbb{R}$ and $\mathbb{C}$, the fundamental and higher homotopy groups of the set of tensors of a fixed small rank, and, taking advantage of Bott periodicity, those of the manifold of tensors of a fixed multilinear rank. We also obtain analogues of these results for symmetric tensors of a fixed symmetric rank or a fixed symmetric multilinear rank.
△ Less
Submitted 21 April, 2018;
originally announced April 2018.
-
A categorification of cyclotomic rings
Authors:
Robert Laugwitz,
You Qi
Abstract:
For any natural number $n \geq 2$, we construct a triangulated monoidal category whose Grothendieck ring is isomorphic to the ring of cyclotomic integers $\mathbb{O}_n$.
For any natural number $n \geq 2$, we construct a triangulated monoidal category whose Grothendieck ring is isomorphic to the ring of cyclotomic integers $\mathbb{O}_n$.
△ Less
Submitted 30 March, 2021; v1 submitted 4 April, 2018;
originally announced April 2018.
-
Morphism spaces in stable categories of Frobenius algebras
Authors:
You Qi
Abstract:
We present an explicit formula computing morphism spaces in the stable category of a Frobenius algebra.
We present an explicit formula computing morphism spaces in the stable category of a Frobenius algebra.
△ Less
Submitted 14 February, 2019; v1 submitted 23 January, 2018;
originally announced January 2018.
-
Limiting Distributions of Spectral Radii for Product of Matrices from the Spherical Ensemble
Authors:
Shuhua Chang,
Deli Li,
Yongcheng Qi
Abstract:
Consider the product of $m$ independent $n\times n$ random matrices from the spherical ensemble for $m\ge 1$. The spectral radius is defined as the maximum absolute value of the $n$ eigenvalues of the product matrix. When $m=1$, the limiting distribution for the spectral radii has been obtained by Jiang and Qi (2017). In this paper, we investigate the limiting distributions for the spectral radii…
▽ More
Consider the product of $m$ independent $n\times n$ random matrices from the spherical ensemble for $m\ge 1$. The spectral radius is defined as the maximum absolute value of the $n$ eigenvalues of the product matrix. When $m=1$, the limiting distribution for the spectral radii has been obtained by Jiang and Qi (2017). In this paper, we investigate the limiting distributions for the spectral radii in general. When $m$ is a fixed integer, we show that the spectral radii converge weakly to distributions of functions of independent Gamma random variables. When $m=m_n$ tends to infinity as $n$ goes to infinity, we show that the logarithmic spectral radii have a normal limit.
△ Less
Submitted 21 January, 2018;
originally announced January 2018.
-
BinaryRelax: A Relaxation Approach For Training Deep Neural Networks With Quantized Weights
Authors:
Penghang Yin,
Shuai Zhang,
Jiancheng Lyu,
Stanley Osher,
Yingyong Qi,
Jack Xin
Abstract:
We propose BinaryRelax, a simple two-phase algorithm, for training deep neural networks with quantized weights. The set constraint that characterizes the quantization of weights is not imposed until the late stage of training, and a sequence of \emph{pseudo} quantized weights is maintained. Specifically, we relax the hard constraint into a continuous regularizer via Moreau envelope, which turns ou…
▽ More
We propose BinaryRelax, a simple two-phase algorithm, for training deep neural networks with quantized weights. The set constraint that characterizes the quantization of weights is not imposed until the late stage of training, and a sequence of \emph{pseudo} quantized weights is maintained. Specifically, we relax the hard constraint into a continuous regularizer via Moreau envelope, which turns out to be the squared Euclidean distance to the set of quantized weights. The pseudo quantized weights are obtained by linearly interpolating between the float weights and their quantizations. A continuation strategy is adopted to push the weights towards the quantized state by gradually increasing the regularization parameter. In the second phase, exact quantization scheme with a small learning rate is invoked to guarantee fully quantized weights. We test BinaryRelax on the benchmark CIFAR and ImageNet color image datasets to demonstrate the superiority of the relaxed quantization approach and the improved accuracy over the state-of-the-art training methods. Finally, we prove the convergence of BinaryRelax under an approximate orthogonality condition.
△ Less
Submitted 4 September, 2018; v1 submitted 19 January, 2018;
originally announced January 2018.
-
Complex best $r$-term approximations almost always exist in finite dimensions
Authors:
Yang Qi,
Mateusz Michałek,
Lek-Heng Lim
Abstract:
We show that in finite-dimensional nonlinear approximations, the best $r$-term approximant of a function $f$ almost always exists over $\mathbb{C}$ but that the same is not true over $\mathbb{R}$, i.e., the infimum $\inf_{f_1,\dots,f_r \in Y} \lVert f - f_1 - \dots - f_r \rVert$ is almost always attainable by complex-valued functions $f_1,\dots, f_r$ in $Y$, a set of functions that have some desir…
▽ More
We show that in finite-dimensional nonlinear approximations, the best $r$-term approximant of a function $f$ almost always exists over $\mathbb{C}$ but that the same is not true over $\mathbb{R}$, i.e., the infimum $\inf_{f_1,\dots,f_r \in Y} \lVert f - f_1 - \dots - f_r \rVert$ is almost always attainable by complex-valued functions $f_1,\dots, f_r$ in $Y$, a set of functions that have some desired structures. Our result extends to functions that possess special properties like symmetry or skew-symmetry under permutations of arguments. For the case where $Y$ is the set of separable functions, the problem becomes that of best rank-$r$ tensor approximations. We show that over $\mathbb{C}$, any tensor almost always has a unique best rank-$r$ approximation. This extends to other notions of tensor ranks such as symmetric rank and alternating rank, to best $r$-block-terms approximations, and to best approximations by tensor networks. When applied to sparse-plus-low-rank approximations, we obtain that for any given $r$ and $k$, a general tensor has a unique best approximation by a sum of a rank-$r$ tensor and a $k$-sparse tensor with a fixed sparsity pattern; this arises in, for example, estimation of covariance matrices of a Gaussian hidden variable model with $k$ observed variables conditionally independent given $r$ hidden variables. The existential (but not the uniqueness) part of our result also applies to best approximations by a sum of a rank-$r$ tensor and a $k$-sparse tensor with no fixed sparsity pattern, as well as to tensor completion problems.
△ Less
Submitted 5 September, 2018; v1 submitted 30 November, 2017;
originally announced November 2017.
-
p-DG cyclotomic nilHecke algebras
Authors:
Mikhail Khovanov,
You Qi,
Joshua Sussan
Abstract:
We categorify a tensor product of two Weyl modules for quantum sl(2) at a prime root of unity.
We categorify a tensor product of two Weyl modules for quantum sl(2) at a prime root of unity.
△ Less
Submitted 1 December, 2021; v1 submitted 20 November, 2017;
originally announced November 2017.
-
Spectral Radii of Truncated Circular Unitary Matrices
Authors:
Wenhao Gui,
Yongcheng Qi
Abstract:
Consider a truncated circular unitary matrix which is a $p_n$ by $p_n$ submatrix of an $n$ by $n$ circular unitary matrix by deleting the last $n-p_n$ columns and rows. Jiang and Qi (2017) proved that the maximum absolute value of the eigenvalues (known as spectral radius) of the truncated matrix, after properly normalized, converges in distribution to the Gumbel distribution if $p_n/n$ is bounded…
▽ More
Consider a truncated circular unitary matrix which is a $p_n$ by $p_n$ submatrix of an $n$ by $n$ circular unitary matrix by deleting the last $n-p_n$ columns and rows. Jiang and Qi (2017) proved that the maximum absolute value of the eigenvalues (known as spectral radius) of the truncated matrix, after properly normalized, converges in distribution to the Gumbel distribution if $p_n/n$ is bounded away from $0$ and $1$. In this paper we investigate the limiting distribution of the spectral radius under one of the following four conditions: (1). $p_n\to\infty$ and $p_n/n\to 0$ as $n\to\infty$; (2). $(n-p_n)/n\to 0$ and $(n-p_n)/(\log n)^3\to\infty$ as $n\to\infty$; (3). $n-p_n\to\infty$ and $(n-p_n)/\log n\to 0$ as $n\to\infty$ and (4). $n-p_n=k\ge 1$ is a fixed integer. We prove that the spectral radius converges in distribution to the Gumbel distribution under the first three conditions and to a reversed Weibull distribution under the fourth condition.
△ Less
Submitted 15 September, 2017;
originally announced September 2017.
-
On Tests for Complete Independence of Normal Random Vectors
Authors:
Shuhua Chang,
Yongcheng Qi
Abstract:
Consider a random sample of $n$ independently and identically distributed $p$-dimensional normal random vectors. A test statistic for complete independence of high-dimensional normal distributions, proposed by Schott (2005), is defined as the sum of squared Pearson's correlation coefficients. A modified test statistic has been proposed by Mao (2014). Under the assumption of complete independence,…
▽ More
Consider a random sample of $n$ independently and identically distributed $p$-dimensional normal random vectors. A test statistic for complete independence of high-dimensional normal distributions, proposed by Schott (2005), is defined as the sum of squared Pearson's correlation coefficients. A modified test statistic has been proposed by Mao (2014). Under the assumption of complete independence, both test statistics are asymptotically normal if the limit $\lim_{n\to\infty}p/n$ exists and is finite. In this paper, we investigate the limiting distributions for both Schott's and Mao's test statistics. We show that both test statistics, after suitably normalized, converge in distribution to the standard normal as long as both $n$ and $p$ tend to infinity. Furthermore, we show that the distribution functions of the test statistics can be approximated very well by a chi-square distribution function with $p(p-1)/2$ degrees of freedom as $n$ tends to infinity regardless of how $p$ changes with $n$.
△ Less
Submitted 5 April, 2017;
originally announced April 2017.
