-
$q$-deformed Howe duality for orthosymplectic Lie superalgebras
Authors:
Jeong Bae,
Jae-Hoon Kwon
Abstract:
We give a $q$-analogue of Howe duality associated to a pair $(\mathfrak{g},G)$, where $\mathfrak{g}$ is an orthosymplectic Lie superalgebra and $G=O_\ell, Sp_{2\ell}$. We define explicitly a commuting action of a quantized enveloping algebra of $\mathfrak{g}$ and the $\imath$quantum group of $\mathfrak{so}_\ell, \mathfrak{sp}_{2\ell}$ on a $q$-deformed supersymmetric space, and describe its semisi…
▽ More
We give a $q$-analogue of Howe duality associated to a pair $(\mathfrak{g},G)$, where $\mathfrak{g}$ is an orthosymplectic Lie superalgebra and $G=O_\ell, Sp_{2\ell}$. We define explicitly a commuting action of a quantized enveloping algebra of $\mathfrak{g}$ and the $\imath$quantum group of $\mathfrak{so}_\ell, \mathfrak{sp}_{2\ell}$ on a $q$-deformed supersymmetric space, and describe its semisimple decomposition whose classical limit recovers the $(\mathfrak{g},G)$-duality. As special cases, we obtain $q$-analogues of $(\mathfrak{g},G)$-dualities on symmetric and exterior algebras for $\mathfrak{g}=\mathfrak{so}_{2n}$, $\mathfrak{sp}_{2n}$.
△ Less
Submitted 6 June, 2025;
originally announced June 2025.
-
Stationary distributions of McKean-Vlasov SDEs with jumps: existence, uniqueness, and multiplicity
Authors:
Jianhai Bao,
Jian Wang
Abstract:
In this paper, we are interested in the issues on existence, uniqueness, and multiplicity of stationary distributions for McKean-Vlasov SDEs with jumps. In detail, with regarding to McKean-Vlasov SDEs driven by pure jump Lévy processes, we principally (i) explore the existence of stationary distributions via Schauder's fixed point theorem under an appropriate Lyapunov condition; (ii) tackle the un…
▽ More
In this paper, we are interested in the issues on existence, uniqueness, and multiplicity of stationary distributions for McKean-Vlasov SDEs with jumps. In detail, with regarding to McKean-Vlasov SDEs driven by pure jump Lévy processes, we principally (i) explore the existence of stationary distributions via Schauder's fixed point theorem under an appropriate Lyapunov condition; (ii) tackle the uniqueness of stationary distributions and the convergence to the equilibria as long as the underlying drifts are continuous with respect to the measure variables under the weighted total variation distance and the $L^1$-Wasserstein distance, respectively; (iii) demonstrate the multiplicity of stationary distributions under a locally dissipative condition. In addition, some illustrative examples are provided to show that the associated McKean-Vlasov SDEs possess a unique, two and three stationary distributions, respectively.
△ Less
Submitted 22 April, 2025;
originally announced April 2025.
-
New Necessary Conditions for Existence of Strong External Difference Families
Authors:
Jingjun Bao,
Lijun Ji
Abstract:
Strong external difference families (SEDFs) were introduced by Paterson and Stinson as a more restrictive version of external difference families. SEDFs can be used to produce optimal strong algebraic manipulation detection codes. In this paper, we use the theory of cyclotomic fields, algebraic number theory and character theory to give some new necessary conditions for the existence of SEDFs. Bas…
▽ More
Strong external difference families (SEDFs) were introduced by Paterson and Stinson as a more restrictive version of external difference families. SEDFs can be used to produce optimal strong algebraic manipulation detection codes. In this paper, we use the theory of cyclotomic fields, algebraic number theory and character theory to give some new necessary conditions for the existence of SEDFs. Based on the results of decomposition of prime ideals and Schmidt's field descent method, two exponent bounds of SEDFs are presented. Based on the field descent method, a special homomorphism from an abelian group to its cyclic subgroup and Gauss sums, some bounds for prime divisors of $v$ and some congruence relations between $k, m$ and $λ$ for $(v,m,k,λ)$-SEDFs with $m>2$ are established.
△ Less
Submitted 15 February, 2025;
originally announced February 2025.
-
Asymptotic expansion at infinity of solutions to Monge-Ampère equation with $C^α$ right term
Authors:
Shuai Qi,
Jiguang Bao
Abstract:
We develop a non-local method to establish the asymptotic expansion at infinity of solutions to Monge-Ampère equation $\det(D^2v)=f$ on $\rn$, where $f$ is a perturbation of $1$ and is only assumed to be Hölder continuous outside a bounded subset of $\rn$, compared to the previous work that $f$ is at least $C^2$.
We develop a non-local method to establish the asymptotic expansion at infinity of solutions to Monge-Ampère equation $\det(D^2v)=f$ on $\rn$, where $f$ is a perturbation of $1$ and is only assumed to be Hölder continuous outside a bounded subset of $\rn$, compared to the previous work that $f$ is at least $C^2$.
△ Less
Submitted 27 January, 2025;
originally announced January 2025.
-
Randomised Euler-Maruyama method for SDEs with Hölder continuous drift coefficient
Authors:
Jianhai Bao,
Yue Wu
Abstract:
In this paper, we examine the performance of randomised Euler-Maruyama (EM) method for additive time-inhomogeneous SDEs with an irregular drift. In particular, the drift is assumed to be $α$-Hölder continuous in time and bounded $β$-Hölder continuous in space with $α,β\in (0,1]$. The strong order of convergence of the randomised EM in $L^p$-norm is shown to be $1/2+(α\wedge (β/2))-ε$ for an arbitr…
▽ More
In this paper, we examine the performance of randomised Euler-Maruyama (EM) method for additive time-inhomogeneous SDEs with an irregular drift. In particular, the drift is assumed to be $α$-Hölder continuous in time and bounded $β$-Hölder continuous in space with $α,β\in (0,1]$. The strong order of convergence of the randomised EM in $L^p$-norm is shown to be $1/2+(α\wedge (β/2))-ε$ for an arbitrary $ε\in (0,1/2)$, higher than the one of standard EM, which is $α\wedge (1/2+β/2-ε)$. The proofs highly rely on the stochastic sewing lemma, where we also provide an alternative proof when handling time irregularity for a comparison.
△ Less
Submitted 26 January, 2025;
originally announced January 2025.
-
The Intrinsic State Variable in Fundamental Lemma and Its Use in Stability Design for Data-based Control
Authors:
Yitao Yan,
Jie Bao,
Biao Huang
Abstract:
In the data-based setting, analysis and control design of dynamical systems using measured data are typically based on overlapping trajectory segments of the input and output variables. This could lead to complex designs because the system internal dynamics, which is typically reflected by the system state variable, is unavailable. In this paper, we will show that the coefficient vector in a modif…
▽ More
In the data-based setting, analysis and control design of dynamical systems using measured data are typically based on overlapping trajectory segments of the input and output variables. This could lead to complex designs because the system internal dynamics, which is typically reflected by the system state variable, is unavailable. In this paper, we will show that the coefficient vector in a modified version of Willems' fundamental lemma is an intrinsic and observable state variable for the system behavior. This argument evolves from the behavioral framework without the requirement of prior knowledge on the causality among system variables or any predefined representation structure (e.g., a state space representation). Such a view allows for the construction of a state map based on the fundamental lemma, bridging the trajectory space and the state space. The state property of the coefficient vector allows for a simple stability design approach using memoryless quadratic functions of it as Lyapunov functions, from which the control action for each step can be explicitly constructed. Using the coefficient vector as a state variable could see wide applications in the analysis and control design of dynamical systems including directions beyond the discussions in this paper.
△ Less
Submitted 21 January, 2025;
originally announced January 2025.
-
Crystals and Double Quiver Algebras from Jeffrey-Kirwan Residues
Authors:
Jiakang Bao,
Masahito Yamazaki
Abstract:
We construct statistical mechanical models of crystal melting describing the flavoured Witten indices of $\mathcal{N}\ge 2$ supersymmetric quiver gauge theories. Our results can be derived from the Jeffrey-Kirwan (JK) residue formulas, and generalize the previous results for quivers corresponding to toric Calabi-Yau threefolds and fourfolds to a large class of quivers satisfying the no-overlap con…
▽ More
We construct statistical mechanical models of crystal melting describing the flavoured Witten indices of $\mathcal{N}\ge 2$ supersymmetric quiver gauge theories. Our results can be derived from the Jeffrey-Kirwan (JK) residue formulas, and generalize the previous results for quivers corresponding to toric Calabi-Yau threefolds and fourfolds to a large class of quivers satisfying the no-overlap condition, including those corresponding to some non-toric Calabi-Yau manifolds. We construct new quiver algebras which we call the double quiver Yangians/algebras, as well as their representations in terms of the aforementioned crystals. For theories with four supercharges, we compare the double quiver algebras with the existing quiver Yangians/BPS algebras, which we show can also be constructed from the JK residues. For theories with two supercharges, the double quiver algebras provide an algebraic description of the BPS states, including the information of the fixed points and their relative coefficients in the full partition functions.
△ Less
Submitted 19 February, 2025; v1 submitted 6 January, 2025;
originally announced January 2025.
-
Atomic Higgsings of 6D SCFTs II: Induced Flows
Authors:
Jiakang Bao,
Hao Y. Zhang
Abstract:
We study a specific type of atomic Higgsings of the 6d $\mathcal{N}=(1,0)$ theories, which we call the induced flows. For the conformal matter theory associated with a pair of nilpotent orbits, the induced flows are given by the inductions of the orbits. We also consider the induced flows for the orbi-instanton theories (as well as some little string theories) that are associated with the homomorp…
▽ More
We study a specific type of atomic Higgsings of the 6d $\mathcal{N}=(1,0)$ theories, which we call the induced flows. For the conformal matter theory associated with a pair of nilpotent orbits, the induced flows are given by the inductions of the orbits. We also consider the induced flows for the orbi-instanton theories (as well as some little string theories) that are associated with the homomorphisms from the discrete subgroups of $\mathrm{SU}(2)$ to $E_8$. This gives a physical definition of the inductions among these discrete homomorphisms, analogous to the inductions of the nilpotent orbits. We analyze the Higgs branch dimensions, the monotonicity of the Weyl anomalies (or the 2-group structure constants for LSTs) and the brane pictures under the induced flows.
△ Less
Submitted 10 April, 2025; v1 submitted 6 January, 2025;
originally announced January 2025.
-
Geometric ergodicity of modified Euler schemes for SDEs with super-linearity
Authors:
Jianhai Bao,
Mateusz B. Majka,
Jian Wang
Abstract:
As a well-known fact, the classical Euler scheme works merely for SDEs with coefficients of linear growth. In this paper, we study a general framework of modified Euler schemes, which is applicable to SDEs with super-linear drifts and encompasses numerical methods such as the tamed Euler scheme and the truncated Euler scheme. On the one hand, by exploiting an approach based on the refined basic co…
▽ More
As a well-known fact, the classical Euler scheme works merely for SDEs with coefficients of linear growth. In this paper, we study a general framework of modified Euler schemes, which is applicable to SDEs with super-linear drifts and encompasses numerical methods such as the tamed Euler scheme and the truncated Euler scheme. On the one hand, by exploiting an approach based on the refined basic coupling, we show that all Euler recursions within our proposed framework are geometrically ergodic under a mixed probability distance (i.e., the total variation distance plus the $L^1$-Wasserstein distance) and the weighted total variation distance. On the other hand, by utilizing the coupling by reflection, we demonstrate that the tamed Euler scheme is geometrically ergodic under the $L^1$-Wasserstein distance. In addition, as an important application, we provide a quantitative $L^1$-Wasserstein error bound between the exact invariant probability measure of an SDE with super-linearity, and the invariant probability measure of the tamed Euler scheme which is its numerical counterpart.
△ Less
Submitted 26 December, 2024;
originally announced December 2024.
-
A note on Lévy-driven McKean-Vlasov SDEs under monotonicity
Authors:
Jianhai Bao,
Yao Liu,
Jian Wang
Abstract:
In this note, under a weak monotonicity and a weak coercivity, we address strong well-posedness of McKean-Vlasov stochastic differential equations (SDEs) driven by Lévy jump processes, where the coefficients are Lipschitz continuous (with respect to the measure variable) under the $L^β$-Wasserstein distance for $β\in[1,2].$ Moreover, the issue on the weak propagation of chaos (i.e., convergence in…
▽ More
In this note, under a weak monotonicity and a weak coercivity, we address strong well-posedness of McKean-Vlasov stochastic differential equations (SDEs) driven by Lévy jump processes, where the coefficients are Lipschitz continuous (with respect to the measure variable) under the $L^β$-Wasserstein distance for $β\in[1,2].$ Moreover, the issue on the weak propagation of chaos (i.e., convergence in distribution via the convergence of the empirical measure) and the strong propagation of chaos (i.e., at the level paths by coupling) is explored simultaneously. To treat the strong well-posedness of McKean-Vlasov SDEs we are interested in, we investigate strong well-posedness of classical time-inhomogeneous SDEs with jumps under a local weak monotonicity and a global weak coercivity. Such a result is of independent interest, and, most importantly, can provide an available reference on strong well-posedness of Lévy-driven SDEs under the monotone condition, which nevertheless is missing for a long time. Based on the theory derived, along with the interlacing technique and the Banach fixed point theorem, the strong well-posedness of McKean-Vlasov SDEs driven by Lévy jump processes can be established. Additionally, as a potential extension, strong well-posedness and conditional propagation of chaos are treated for Lévy-driven McKean-Vlasov SDEs with common noise under a weak monotonicity.
△ Less
Submitted 1 December, 2024;
originally announced December 2024.
-
$L^2$-Wasserstein contraction of modified Euler schemes for SDEs with high diffusivity and applications
Authors:
Jianhai Bao,
Jiaqing Hao
Abstract:
In this paper, we are concerned with a modified Euler scheme for the SDE under consideration, where the drift is of super-linear growth and dissipative merely outside a closed ball. By adopting the synchronous coupling, along with the construction of an equivalent quasi-metric, the $L^2$-Wasserstein contraction of the modified Euler scheme is addressed provided that the diffusivity is large enough…
▽ More
In this paper, we are concerned with a modified Euler scheme for the SDE under consideration, where the drift is of super-linear growth and dissipative merely outside a closed ball. By adopting the synchronous coupling, along with the construction of an equivalent quasi-metric, the $L^2$-Wasserstein contraction of the modified Euler scheme is addressed provided that the diffusivity is large enough. In particular, as a by-product, the $L^2$ Wasserstein contraction of the projected (truncated) Euler scheme and the tamed Euler algorithm is treated under much more explicit conditions imposed on drifts. The theory derived on the $L^2$-Wasserstein contraction has numerous applications on various aspects. In addition to applications on Poincaré inequalities (with respect to the numerical transition kernel and the numerical invariant probability measure), concentration inequalities for empirical averages, and bounds concerning the KL-divergence, in this paper we present another two potential applications. One concerns the non-asymptotic $L^2$-Wasserstein bound corresponding to the projected Euler scheme and the tamed Euler recursion, respectively, which further implies the $L^2$-Wasserstein error bound between the exact invariant probability measure and the numerical counterpart. It is worthy to emphasize that the associated convergence rate is improved greatly in contrast to the existing literature. Another application is devoted to the strong law of large numbers of additive functionals related to the modified Euler algorithm, where the observable functions involved are allowed to be of polynomial growth, and the associated convergence rate is also enhanced remarkably.
△ Less
Submitted 3 November, 2024;
originally announced November 2024.
-
Futaki Invariants and Reflexive Polygons
Authors:
Jiakang Bao,
Eugene Choi,
Yang-Hui He,
Rak-Kyeong Seong,
Shing-Tung Yau
Abstract:
Futaki invariants of the classical moduli space of 4d N=1 supersymmetric gauge theories determine whether they have a conformal fixed point in the IR. We systematically compute the Futaki invariants for a large family of 4d N=1 supersymmetric gauge theories coming from D3-branes probing Calabi-Yau 3-fold singularities whose bases are Gorenstein Fano surfaces. In particular, we focus on the toric c…
▽ More
Futaki invariants of the classical moduli space of 4d N=1 supersymmetric gauge theories determine whether they have a conformal fixed point in the IR. We systematically compute the Futaki invariants for a large family of 4d N=1 supersymmetric gauge theories coming from D3-branes probing Calabi-Yau 3-fold singularities whose bases are Gorenstein Fano surfaces. In particular, we focus on the toric case where the Fano surfaces are given by the 16 reflexive convex polygons and the moduli spaces are given by the corresponding toric Calabi-Yau 3-folds. We study the distribution of and conjecture new bounds on the Futaki invariants with respect to various topological and geometric quantities. These include the minimum volume of the Sasaki-Einstein base manifolds as well as the Chern and Euler numbers of the toric Fano surfaces. Even though the moduli spaces for the family of theories studied are known to be K-stable, our work sheds new light on how the topological and geometric quantities restrict the Futaki invariants for a plethora of moduli spaces.
△ Less
Submitted 24 October, 2024;
originally announced October 2024.
-
Emergence of peaked singularities in the Euler-Poisson system
Authors:
Junsik Bae,
Sang-Hyuck Moon,
Kwan Woo
Abstract:
We consider the one-dimensional Euler-Poisson system equipped with the Boltzmann relation and provide the exact asymptotic behavior of the peaked solitary wave solutions near the peak. This enables us to study the cold ion limit of the peaked solitary waves with the sharp range of Hölder exponents. Furthermore, we provide numerical evidence for $C^1$ blow-up solutions to the pressureless Euler-Poi…
▽ More
We consider the one-dimensional Euler-Poisson system equipped with the Boltzmann relation and provide the exact asymptotic behavior of the peaked solitary wave solutions near the peak. This enables us to study the cold ion limit of the peaked solitary waves with the sharp range of Hölder exponents. Furthermore, we provide numerical evidence for $C^1$ blow-up solutions to the pressureless Euler-Poisson system, whose blow-up profiles are asymptotically similar to its peaked solitary waves and exhibit a different form of blow-up compared to the Burgers-type (shock-like) blow-up.
△ Less
Submitted 12 September, 2024;
originally announced September 2024.
-
The random periodic solutions for McKean-Vlasov stochastic differential equations
Authors:
Jianhai Bao,
Goncalo Dos Reis,
Yue Wu
Abstract:
In this paper, we study well-posedness of random periodic solutions of stochastic differential equations (SDEs) of McKean-Vlasov type driven by a two-sided Brownian motion, where the random periodic behaviour is characterised by the equations' long-time behaviour. Given the well-known connection between McKean-Vlasov SDEs and interacting particle systems, we show propagation of chaos and that the…
▽ More
In this paper, we study well-posedness of random periodic solutions of stochastic differential equations (SDEs) of McKean-Vlasov type driven by a two-sided Brownian motion, where the random periodic behaviour is characterised by the equations' long-time behaviour. Given the well-known connection between McKean-Vlasov SDEs and interacting particle systems, we show propagation of chaos and that the key properties of the interacting particle systems recover those of the McKean-Vlasov SDEs in the particle limit. All results in the present work are shown under two settings: fully and partially dissipative case. Each setting has its challenges and limitations. For instance, weakening full dissipativity to partial dissipativity demands stronger structural assumptions on the equations' dynamics and yields random periodic behaviour in the weak sense instead of pathwise sense (as in the full dissipativity case). The proof mechanisms are close but fundamentally different.
△ Less
Submitted 4 December, 2024; v1 submitted 30 August, 2024;
originally announced August 2024.
-
Singularity formation of hydromagnetic waves in cold plasma
Authors:
Junsik Bae,
Junho Choi,
Bongsuk Kwon
Abstract:
We study $C^1$ blow-up of the compressible fluid model introduced by Gardner and Morikawa, which describes the dynamics of a magnetized cold plasma. We propose sufficient conditions that lead to $C^1$ blow-up. In particular, we find that smooth solutions can break down in finite time even if the gradient of initial velocity is identically zero. The density and the gradient of the velocity become u…
▽ More
We study $C^1$ blow-up of the compressible fluid model introduced by Gardner and Morikawa, which describes the dynamics of a magnetized cold plasma. We propose sufficient conditions that lead to $C^1$ blow-up. In particular, we find that smooth solutions can break down in finite time even if the gradient of initial velocity is identically zero. The density and the gradient of the velocity become unbounded as time approaches the lifespan of the smooth solution. The Lagrangian formulation reduces the singularity formation problem to finding a zero of the associated second-order ODE.
△ Less
Submitted 26 July, 2024;
originally announced July 2024.
-
Delta-shock for the pressureless Euler-Poisson system
Authors:
Junsik Bae,
Yunjoo Kim,
Bongsuk Kwon
Abstract:
We study singularity formation for the pressureless Euler-Poisson system of cold ion dynamics. In contrast to the Euler-Poisson system with pressure, when its smooth solutions experience $C^1$ blow-up, the $L^\infty$ norm of the density becomes unbounded, which is often referred to as a delta-shock. We provide a constructive proof of singularity formation to obtain an exact blow-up profile and the…
▽ More
We study singularity formation for the pressureless Euler-Poisson system of cold ion dynamics. In contrast to the Euler-Poisson system with pressure, when its smooth solutions experience $C^1$ blow-up, the $L^\infty$ norm of the density becomes unbounded, which is often referred to as a delta-shock. We provide a constructive proof of singularity formation to obtain an exact blow-up profile and the detailed asymptotic behavior of the solutions near the blow-up point in both time and space. Our result indicates that at the blow-up time $t=T_\ast$, the density function is unbounded but is locally integrable with the profile of $ρ(x,T_\ast) \sim (x-x_*)^{-2/3}$ near the blow-up point $x=x_\ast$. This profile is not yet a Dirac measure. On the other hand, the velocity function has $C^{1/3}$ regularity at the blow-up point. Loosely following our analysis, we also obtain an exact blow-up profile for the pressureless Euler equations.
△ Less
Submitted 22 July, 2024;
originally announced July 2024.
-
Partial Difference Sets with Denniston Parameters in Elementary Abelian $p$-Groups
Authors:
Jingjun Bao,
Qing Xiang,
Meng Zhao
Abstract:
Denniston \cite{D1969} constructed partial difference sets (PDS) with parameters $(2^{3m}, (2^{m+r}-2^m+2^r)(2^m-1), 2^m-2^r+(2^{m+r}-2^m+2^r)(2^r-2), (2^{m+r}-2^m+2^r)(2^r-1))$ in elementary abelian groups of order $2^{3m}$ for all $m\geq 2$ and $1 \leq r < m$. These PDS correspond to maximal arcs in the Desarguesian projective planes PG$(2, 2^m)$. Davis et al. \cite{DHJP2024} and also De Winter…
▽ More
Denniston \cite{D1969} constructed partial difference sets (PDS) with parameters $(2^{3m}, (2^{m+r}-2^m+2^r)(2^m-1), 2^m-2^r+(2^{m+r}-2^m+2^r)(2^r-2), (2^{m+r}-2^m+2^r)(2^r-1))$ in elementary abelian groups of order $2^{3m}$ for all $m\geq 2$ and $1 \leq r < m$. These PDS correspond to maximal arcs in the Desarguesian projective planes PG$(2, 2^m)$. Davis et al. \cite{DHJP2024} and also De Winter \cite{dewinter23} presented constructions of PDS with Denniston parameters $(p^{3m}, (p^{m+r}-p^m+p^r)(p^m-1), p^m-p^r+(p^{m+r}-p^m+p^r)(p^r-2), (p^{m+r}-p^m+p^r)(p^r-1))$ in elementary abelian groups of order $p^{3m}$ for all $m \geq 2$ and $r \in \{1, m-1\}$, where $p$ is an odd prime. The constructions in \cite{DHJP2024, dewinter23} are particularly intriguing, as it was shown by Ball, Blokhuis, and Mazzocca \cite{BBM1997} that no nontrivial maximal arcs in PG$(2, q^m)$ exist for any odd prime power $q$. In this paper, we show that PDS with Denniston parameters $(q^{3m}, (q^{m+r}-q^m+q^r)(q^m-1), q^m-q^r+(q^{m+r}-q^m+q^r)(q^r-2), (q^{m+r}-q^m+q^r)(q^r-1))$ exist in elementary abelian groups of order $q^{3m}$ for all $m \geq 2$ and $1 \leq r < m$, where $q$ is an arbitrary prime power.
△ Less
Submitted 22 July, 2024;
originally announced July 2024.
-
Uniform-in-time estimates for mean-field type SDEs and applications
Authors:
Jianhai Bao,
Jiaqing Hao
Abstract:
Via constructing an asymptotic coupling by reflection, in this paper we establish uniform-in-time estimates on probability distances for mean-field type SDEs, where the drift terms under consideration are dissipative merely in the long distance. As applications, we (i) explore the long time probability distance estimate between an SDE and its delay version; (ii) investigate the issue on uniform-in…
▽ More
Via constructing an asymptotic coupling by reflection, in this paper we establish uniform-in-time estimates on probability distances for mean-field type SDEs, where the drift terms under consideration are dissipative merely in the long distance. As applications, we (i) explore the long time probability distance estimate between an SDE and its delay version; (ii) investigate the issue on uniform-in-time propagation of chaos for McKean-Vlasov SDEs, where the drifts might be singular with respect to the spatial variables and need not to be of convolution type; (iii) tackle the discretization error bounds in an infinite-time horizon for stochastic algorithms (e.g. backward/tamed/adaptive Euler-Maruyama schemes as three typical candidates) associated with McKean-Vlasov SDEs.
△ Less
Submitted 25 September, 2024; v1 submitted 7 May, 2024;
originally announced May 2024.
-
Structure of singularities for the Euler-Poisson system of ion dynamics
Authors:
Junsik Bae,
Yunjoo Kim,
Bongsuk Kwon
Abstract:
We study the formation of singularity for the isothermal Euler-Poisson system arising from plasma physics. Contrast to the previous studies yielding only limited information on the blow-up solutions, for instance, sufficient conditions for the blow-up and the temporal blow-up rate along the characteristic curve, we rather give a constructive proof of singularity formation from smooth initial data.…
▽ More
We study the formation of singularity for the isothermal Euler-Poisson system arising from plasma physics. Contrast to the previous studies yielding only limited information on the blow-up solutions, for instance, sufficient conditions for the blow-up and the temporal blow-up rate along the characteristic curve, we rather give a constructive proof of singularity formation from smooth initial data. More specifically, employing the stable blow-up profile of the Burgers equation in the self-similar variables, we establish the global stability estimate in the self-similar time, which yields the asymptotic behavior of blow-up solutions near the singularity point. Our analysis indicates that the smooth solution to the Euler-Poisson system can develop a cusp-type singularity; it exhibits $C^1$ blow-up in a finite time, while it belongs to $C^{1/3}$ at the blow-up time, provided that smooth initial data are sufficiently close to the blow-up profile in some weighted $C^4$-topology. We also present a similar result for the isentropic case, and discuss noteworthy differences in the analysis.
△ Less
Submitted 4 May, 2024;
originally announced May 2024.
-
On the Reachability of 3-Dimensional Paths with a Prescribed Curvature Bound
Authors:
Juho Bae,
Ji Hoon Bai,
Byung-Yoon Lee,
Jun-Yong Lee,
Chang-Hun Lee
Abstract:
This paper presents the reachability analysis of curves in $\mathbb{R}^3$ with a prescribed curvature bound. Based on Pontryagin Maximum Principle, we leverage the existing knowledge on the structure of solutions to minimum-time problems, or Markov-Dubins problem, to reachability considerations. Based on this development, two types of reachability are discussed. First, we prove that any boundary p…
▽ More
This paper presents the reachability analysis of curves in $\mathbb{R}^3$ with a prescribed curvature bound. Based on Pontryagin Maximum Principle, we leverage the existing knowledge on the structure of solutions to minimum-time problems, or Markov-Dubins problem, to reachability considerations. Based on this development, two types of reachability are discussed. First, we prove that any boundary point of the reachability set, with the directional component taken into account as well as geometric coordinates, can be reached via curves of H, CSC, CCC, or their respective subsegments, where H denotes a helicoidal arc, C a circular arc with maximum curvature, and S a straight segment. Second, we show that the reachability set when directional component is not considered\textemdash{}the position reachability set\textemdash{}is simply a solid of revolution of its two-dimensional counterpart, the Dubins car. These findings extend the developments presented in literature on Dubins car into spatial curves in $\mathbb{R}^3$.
△ Less
Submitted 26 March, 2025; v1 submitted 27 March, 2024;
originally announced March 2024.
-
Limit theorems for SDEs with irregular drifts
Authors:
Jianhai Bao,
Jiaqing Hao
Abstract:
In this paper, concerning SDEs with Hölder continuous drifts, which are merely dissipative at infinity, and SDEs with piecewise continuous drifts, we investigate the strong law of large numbers and the central limit theorem for underlying additive functionals and reveal the corresponding rates of convergence. To establish the limit theorems under consideration, the exponentially contractive proper…
▽ More
In this paper, concerning SDEs with Hölder continuous drifts, which are merely dissipative at infinity, and SDEs with piecewise continuous drifts, we investigate the strong law of large numbers and the central limit theorem for underlying additive functionals and reveal the corresponding rates of convergence. To establish the limit theorems under consideration, the exponentially contractive property of solution processes under the (quasi-)Wasserstein distance plays an indispensable role. In order to achieve such contractive property, which is new and interesting in its own right for SDEs with Hölder continuous drifts or piecewise continuous drifts, the reflection coupling method is employed and meanwhile a sophisticated test function is built.
△ Less
Submitted 10 March, 2024;
originally announced March 2024.
-
Can We Remove the Square-Root in Adaptive Gradient Methods? A Second-Order Perspective
Authors:
Wu Lin,
Felix Dangel,
Runa Eschenhagen,
Juhan Bae,
Richard E. Turner,
Alireza Makhzani
Abstract:
Adaptive gradient optimizers like Adam(W) are the default training algorithms for many deep learning architectures, such as transformers. Their diagonal preconditioner is based on the gradient outer product which is incorporated into the parameter update via a square root. While these methods are often motivated as approximate second-order methods, the square root represents a fundamental differen…
▽ More
Adaptive gradient optimizers like Adam(W) are the default training algorithms for many deep learning architectures, such as transformers. Their diagonal preconditioner is based on the gradient outer product which is incorporated into the parameter update via a square root. While these methods are often motivated as approximate second-order methods, the square root represents a fundamental difference. In this work, we investigate how the behavior of adaptive methods changes when we remove the root, i.e., strengthen their second-order motivation. Surprisingly, we find that such square-root-free adaptive methods close the generalization gap to SGD on convolutional architectures, while maintaining their root-based counterpart's performance on transformers. The second-order perspective also has practical benefits for developing non-diagonal methods that can incorporate arbitrary curvature approximations through the concept of preconditioner invariance. In contrast to root-based methods like Shampoo, root-free counterparts work well and fast with half-precision since they do not require numerically unstable matrix root decompositions and inversions. Overall, our findings provide new insights into the development of adaptive methods and raise important questions regarding the overlooked role of adaptivity in their success. (experiment code: https://github.com/yorkerlin/remove-the-square-root optimizer code: https://github.com/f-dangel/sirfshampoo)
△ Less
Submitted 4 October, 2024; v1 submitted 5 February, 2024;
originally announced February 2024.
-
Two Necessary and Sufficient Conditions to the Solvability of the Exterior Dirichlet Problem for the Monge-Ampère Equation
Authors:
Cong Wang,
Jiguang Bao
Abstract:
The present paper provides two necessary and sufficient conditions for the existence of solutions to the exterior Dirichlet problem of the Monge-Ampère equation with prescribed asymptotic behavior at infinity. By an adapted smooth approximation argument, we prove that the problem is solvable if and only if the boundary value is semi-convex with respect to the inner boundary, which is our first pro…
▽ More
The present paper provides two necessary and sufficient conditions for the existence of solutions to the exterior Dirichlet problem of the Monge-Ampère equation with prescribed asymptotic behavior at infinity. By an adapted smooth approximation argument, we prove that the problem is solvable if and only if the boundary value is semi-convex with respect to the inner boundary, which is our first proposed new concept. Along the lines of Perron's method for Laplace equation, we obtain the threshold for solvability in the asymptotic behavior at infinity of the solution, and remove the $C^2$ regularity assumptions on the boundary value and on the inner boundary which are required in the proofs of the corresponding existence theorems in the recent literatures.
△ Less
Submitted 22 January, 2024;
originally announced January 2024.
-
The Origin of Calabi-Yau Crystals in BPS States Counting
Authors:
Jiakang Bao,
Rak-Kyeong Seong,
Masahito Yamazaki
Abstract:
We study the counting problem of BPS D-branes wrapping holomorphic cycles of a general toric Calabi-Yau manifold. We evaluate the Jeffrey-Kirwan residues for the flavoured Witten index for the supersymmetric quiver quantum mechanics on the worldvolume of the D-branes, and find that BPS degeneracies are described by a statistical mechanical model of crystal melting. For Calabi-Yau threefolds, we re…
▽ More
We study the counting problem of BPS D-branes wrapping holomorphic cycles of a general toric Calabi-Yau manifold. We evaluate the Jeffrey-Kirwan residues for the flavoured Witten index for the supersymmetric quiver quantum mechanics on the worldvolume of the D-branes, and find that BPS degeneracies are described by a statistical mechanical model of crystal melting. For Calabi-Yau threefolds, we reproduce the crystal melting models long known in the literature. For Calabi-Yau fourfolds, however, we find that the crystal does not contain the full information for the BPS degeneracy and we need to explicitly evaluate non-trivial weights assigned to the crystal configurations. Our discussions treat Calabi-Yau threefolds and fourfolds on equal footing, and include discussions on elliptic and rational generalizations of the BPS states counting, connections to the mathematical definition of generalized Donaldson-Thomas invariants, examples of wall crossings, and of trialities in quiver gauge theories.
△ Less
Submitted 12 March, 2025; v1 submitted 5 January, 2024;
originally announced January 2024.
-
Rabinowitz Floer homology for prequantization bundles and Floer Gysin sequence
Authors:
Joonghyun Bae,
Jungsoo Kang,
Sungho Kim
Abstract:
Let $Y$ be a prequantization bundle over a closed spherically monotone symplectic manifold $Σ$. Adapting an idea due to Diogo and Lisi, we study a split version of Rabinowitz Floer homology for $Y$ in the following two settings. First, $Σ$ is a symplectic hyperplane section of a closed symplectic manifold $X$ satisfying a certain monotonicity condition; in this case, $X \setminus Σ$ is a Liouville…
▽ More
Let $Y$ be a prequantization bundle over a closed spherically monotone symplectic manifold $Σ$. Adapting an idea due to Diogo and Lisi, we study a split version of Rabinowitz Floer homology for $Y$ in the following two settings. First, $Σ$ is a symplectic hyperplane section of a closed symplectic manifold $X$ satisfying a certain monotonicity condition; in this case, $X \setminus Σ$ is a Liouville filling of $Y$. Second, the minimal Chern number of $Σ$ is greater than one, which is the case where the Rabinowitz Floer homology of the symplectization $\mathbb{R} \times Y$ is defined. In both cases, we construct a Gysin-type exact sequence connecting the Rabinowitz Floer homology of $X\setminusΣ$ or $\mathbb{R} \times Y$ and the quantum homology of $Σ$. As applications, we discuss the invertibility of a symplectic hyperplane section class in quantum homology, the isotopy problem for fibered Dehn twists, the orderability problem for prequantization bundles, and the existence of translated points. We also provide computational results based on the exact sequence that we construct.
△ Less
Submitted 9 April, 2024; v1 submitted 29 November, 2023;
originally announced November 2023.
-
Existence and asymptotic behavior of entire large solutions for Hessian equations
Authors:
Xiang Li,
Jiguang Bao
Abstract:
In this paper, we give some existence and nonexistence results for nonradial entire large solutions of the Hessian equation $S_k\left(D^2 u\right)=b(x) u^γ$ in the sublinear case $0<γ<k$. The exact asymptotic behavior of large solutions at infinity is also studied when $b(x)$ is the oscillation of a radial function $|x|^{-l}$ at infinity for $l\leq k-1$.
In this paper, we give some existence and nonexistence results for nonradial entire large solutions of the Hessian equation $S_k\left(D^2 u\right)=b(x) u^γ$ in the sublinear case $0<γ<k$. The exact asymptotic behavior of large solutions at infinity is also studied when $b(x)$ is the oscillation of a radial function $|x|^{-l}$ at infinity for $l\leq k-1$.
△ Less
Submitted 14 October, 2023;
originally announced October 2023.
-
Ensemble Kalman Filters with Resampling
Authors:
Omar Al Ghattas,
Jiajun Bao,
Daniel Sanz-Alonso
Abstract:
Filtering is concerned with online estimation of the state of a dynamical system from partial and noisy observations. In applications where the state of the system is high dimensional, ensemble Kalman filters are often the method of choice. These algorithms rely on an ensemble of interacting particles to sequentially estimate the state as new observations become available. Despite the practical su…
▽ More
Filtering is concerned with online estimation of the state of a dynamical system from partial and noisy observations. In applications where the state of the system is high dimensional, ensemble Kalman filters are often the method of choice. These algorithms rely on an ensemble of interacting particles to sequentially estimate the state as new observations become available. Despite the practical success of ensemble Kalman filters, theoretical understanding is hindered by the intricate dependence structure of the interacting particles. This paper investigates ensemble Kalman filters that incorporate an additional resampling step to break the dependency between particles. The new algorithm is amenable to a theoretical analysis that extends and improves upon those available for filters without resampling, while also performing well in numerical examples.
△ Less
Submitted 27 July, 2024; v1 submitted 16 August, 2023;
originally announced August 2023.
-
Nonexistence of multi-dimensional solitary waves for the Euler-Poisson system
Authors:
Junsik Bae,
Daisuke Kawagoe
Abstract:
We study the nonexistence of multi-dimensional solitary waves for the Euler-Poisson system governing ion dynamics. It is well-known that the one-dimensional Euler-Poisson system has solitary waves that travel faster than the ion-sound speed. In contrast, we show that the two-dimensional and three-dimensional models do not admit nontrivial irrotational spatially localized traveling waves for any tr…
▽ More
We study the nonexistence of multi-dimensional solitary waves for the Euler-Poisson system governing ion dynamics. It is well-known that the one-dimensional Euler-Poisson system has solitary waves that travel faster than the ion-sound speed. In contrast, we show that the two-dimensional and three-dimensional models do not admit nontrivial irrotational spatially localized traveling waves for any traveling velocity and for general pressure laws. We derive some Pohozaev type identities associated with the energy and density integrals. This approach is extended to prove the nonexistence of irrotational multi-dimensional solitary waves for the two-species Euler-Poisson system for ions and electrons.
△ Less
Submitted 7 August, 2023;
originally announced August 2023.
-
Interior derivative estimates and Bernstein theorem for Hessian quotient equations
Authors:
Limei Dai,
Jiguang Bao,
Bo Wang
Abstract:
In this paper, we obtain the interior derivative estimates of solutions for elliptic and parabolic Hessian quotient equations. Then we establish the Bernstein theorem for parabolic Hessian quotient equations, that is, any parabolically convex solution $u=u(x,t)\in C^{4,2}(\mathbb{R}^n\times (-\infty,0])$ for $-u_t\frac{S_n(D^2u)}{S_l(D^2u)}=1$ in $\mathbb{R}^n\times (-\infty,0]$ must be the form o…
▽ More
In this paper, we obtain the interior derivative estimates of solutions for elliptic and parabolic Hessian quotient equations. Then we establish the Bernstein theorem for parabolic Hessian quotient equations, that is, any parabolically convex solution $u=u(x,t)\in C^{4,2}(\mathbb{R}^n\times (-\infty,0])$ for $-u_t\frac{S_n(D^2u)}{S_l(D^2u)}=1$ in $\mathbb{R}^n\times (-\infty,0]$ must be the form of $u=-mt+P(x)$ with $m>0$ being a constant and $P$ being a convex quadratic polynomial.
△ Less
Submitted 28 May, 2023;
originally announced May 2023.
-
Entire solutions to the parabolic Monge--Ampère equation with unbounded nonlinear growth in time
Authors:
Ning An,
Jiguang Bao,
Zixiao Liu
Abstract:
The Liouville type theorem on the parabolic Monge--Ampère equation $-u_t\det D^2u=1$ states that any entire parabolically convex classical solution must be of form $-t+|x|^2/2$ up to a re-scaling and transformation, under additional assumption that partial derivative with respect to time variable $u_t$ is strictly negative and bounded. In this paper, we study the case when $u_t$ is unbounded, prov…
▽ More
The Liouville type theorem on the parabolic Monge--Ampère equation $-u_t\det D^2u=1$ states that any entire parabolically convex classical solution must be of form $-t+|x|^2/2$ up to a re-scaling and transformation, under additional assumption that partial derivative with respect to time variable $u_t$ is strictly negative and bounded. In this paper, we study the case when $u_t$ is unbounded, prove an existence result of entire parabolically convex smooth solution and investigate the asymptotic behavior near infinity.
△ Less
Submitted 14 May, 2023;
originally announced May 2023.
-
Scientific Computing Algorithms to Learn Enhanced Scalable Surrogates for Mesh Physics
Authors:
Brian R. Bartoldson,
Yeping Hu,
Amar Saini,
Jose Cadena,
Yucheng Fu,
Jie Bao,
Zhijie Xu,
Brenda Ng,
Phan Nguyen
Abstract:
Data-driven modeling approaches can produce fast surrogates to study large-scale physics problems. Among them, graph neural networks (GNNs) that operate on mesh-based data are desirable because they possess inductive biases that promote physical faithfulness, but hardware limitations have precluded their application to large computational domains. We show that it is \textit{possible} to train a cl…
▽ More
Data-driven modeling approaches can produce fast surrogates to study large-scale physics problems. Among them, graph neural networks (GNNs) that operate on mesh-based data are desirable because they possess inductive biases that promote physical faithfulness, but hardware limitations have precluded their application to large computational domains. We show that it is \textit{possible} to train a class of GNN surrogates on 3D meshes. We scale MeshGraphNets (MGN), a subclass of GNNs for mesh-based physics modeling, via our domain decomposition approach to facilitate training that is mathematically equivalent to training on the whole domain under certain conditions. With this, we were able to train MGN on meshes with \textit{millions} of nodes to generate computational fluid dynamics (CFD) simulations. Furthermore, we show how to enhance MGN via higher-order numerical integration, which can reduce MGN's error and training time. We validated our methods on an accompanying dataset of 3D $\text{CO}_2$-capture CFD simulations on a 3.1M-node mesh. This work presents a practical path to scaling MGN for real-world applications.
△ Less
Submitted 1 April, 2023;
originally announced April 2023.
-
Existence of entire solutions to the Lagrangian mean curvature equations in supercritical phase
Authors:
Zixiao Liu,
Cong Wang,
Jiguang Bao
Abstract:
In this paper, we establish the existence and uniqueness theorem of entire solutions to the Lagrangian mean curvature equations with prescribed asymptotic behavior at infinity. The phase functions are assumed to be supercritical and converge to a constant in a certain rate at infinity. The basic idea is to establish uniform estimates for the approximating problems defined on bounded domains and th…
▽ More
In this paper, we establish the existence and uniqueness theorem of entire solutions to the Lagrangian mean curvature equations with prescribed asymptotic behavior at infinity. The phase functions are assumed to be supercritical and converge to a constant in a certain rate at infinity. The basic idea is to establish uniform estimates for the approximating problems defined on bounded domains and the main ingredient is to construct appropriate subsolutions and supersolutions as barrier functions. We also prove a nonexistence result to show the convergence rate of the phase functions is optimal.
△ Less
Submitted 14 February, 2023;
originally announced February 2023.
-
Interior estimates of derivatives and a Liouville type theorem for Parabolic $k$-Hessian equations
Authors:
Jiguang Bao,
Jiechen Qiang,
Zhongwei Tang,
Cong Wang
Abstract:
In this paper, we establish the gradient and Pogorelov estimates for $k$-convex-monotone solutions to parabolic $k$-Hessian equations of the form $-u_tσ_k(λ(D^2u))=ψ(x,t,u)$. We also apply such estimates to obtain a Liouville type result, which states that any $k$-convex-monotone and $C^{4,2}$ solution $u$ to $-u_tσ_k(λ(D^2u))=1$ in $\mathbb{R}^n\times(-\infty,0]$ must be a linear function of $t$…
▽ More
In this paper, we establish the gradient and Pogorelov estimates for $k$-convex-monotone solutions to parabolic $k$-Hessian equations of the form $-u_tσ_k(λ(D^2u))=ψ(x,t,u)$. We also apply such estimates to obtain a Liouville type result, which states that any $k$-convex-monotone and $C^{4,2}$ solution $u$ to $-u_tσ_k(λ(D^2u))=1$ in $\mathbb{R}^n\times(-\infty,0]$ must be a linear function of $t$ plus a quadratic polynomial of $x$, under some growth assumptions on $u$.
△ Less
Submitted 13 January, 2023; v1 submitted 22 September, 2022;
originally announced September 2022.
-
Normalized solutions to lower critical Choquard equation with a local perturbation
Authors:
Xinfu Li,
Jianguang Bao,
Wenguang Tang
Abstract:
In this paper, we study the existence and non-existence of normalized solutions to the lower critical Choquard equation with a local perturbation \begin{equation*} \begin{cases} -Δu+λu=γ(I_α\ast|u|^{\frac{N+α}{N}})|u|^{\frac{N+α}{N}-2}u+μ|u|^{q-2}u,\quad \text{in}\ \mathbb{R}^N, \\ \int_{\mathbb{R}^N}|u|^2dx=c^2, \end{cases} \end{equation*} where $γ, μ, c>0$, $2<q\leq 2+\frac{4}{N}$, and…
▽ More
In this paper, we study the existence and non-existence of normalized solutions to the lower critical Choquard equation with a local perturbation \begin{equation*} \begin{cases} -Δu+λu=γ(I_α\ast|u|^{\frac{N+α}{N}})|u|^{\frac{N+α}{N}-2}u+μ|u|^{q-2}u,\quad \text{in}\ \mathbb{R}^N, \\ \int_{\mathbb{R}^N}|u|^2dx=c^2, \end{cases} \end{equation*} where $γ, μ, c>0$, $2<q\leq 2+\frac{4}{N}$, and $λ\in\mathbb{R}$ is an unknown parameter that appears as a Lagrange multiplier. The results of this paper about this equation answer some questions proposed by Yao, Chen, Rǎdulescu and Sun [Siam J. Math. Anal., 54(3) (2022), 3696-3723]. Moreover, based on the results obtained, we study the multiplicity of normalized solutions to the non-autonomous Choquard equation \begin{equation*} \begin{cases} -Δu+λu=(I_α\ast [h(εx)|u|^{\frac{N+α}{N}}])h(εx)|u|^{\frac{N+α}{N}-2}u+μ|u|^{q-2}u,\ x\in \mathbb{R}^N, \\ \int_{\mathbb{R}^N}|u|^2dx=c^2, \end{cases} \end{equation*} where $ε>0$, $2<q<2+\frac{4}{N}$, and $h$ is a positive and continuous function. It is proved that the numbers of normalized solutions are at least the numbers of global maximum points of $h$ when $ε$ is small enough.
△ Less
Submitted 18 August, 2022; v1 submitted 21 July, 2022;
originally announced July 2022.
-
Necessary and sufficient conditions on existence of radial solutions for exterior Dirichlet problem of fully nonlinear elliptic equations
Authors:
Limei Dai,
Jiguang Bao,
Bo Wang
Abstract:
In this paper, we study the exterior Dirichlet problem for the fully nonlinear elliptic equation $f(λ(D^{2}u))=1$. We obtain the necessary and sufficient conditions of existence of radial solutions with prescribed asymptotic behavior at infinity.
In this paper, we study the exterior Dirichlet problem for the fully nonlinear elliptic equation $f(λ(D^{2}u))=1$. We obtain the necessary and sufficient conditions of existence of radial solutions with prescribed asymptotic behavior at infinity.
△ Less
Submitted 17 June, 2022;
originally announced June 2022.
-
The exterior Dirichlet problems of Hessian quotient equations
Authors:
Limei Dai,
Jiguang Bao,
Bo Wang
Abstract:
In this paper, we study the Dirichlet problem of Hessian quotient equations in exterior domains. By estimating the eigenvalues of the solution, the necessary and sufficient conditions on existence of radial solutions are obtained. Applying the solutions of ODE, the viscosity subsolutions and supersolutions are constructed and then the existence of viscosity solutions for exterior problems is estab…
▽ More
In this paper, we study the Dirichlet problem of Hessian quotient equations in exterior domains. By estimating the eigenvalues of the solution, the necessary and sufficient conditions on existence of radial solutions are obtained. Applying the solutions of ODE, the viscosity subsolutions and supersolutions are constructed and then the existence of viscosity solutions for exterior problems is established by the Perron's method.
△ Less
Submitted 17 June, 2022;
originally announced June 2022.
-
A Note on Quiver Yangians and $\mathcal{R}$-Matrices
Authors:
Jiakang Bao
Abstract:
In this note, we study possible $\mathcal{R}$-matrix constructions in the context of quiver Yangians and Yang-Baxter algebras. For generalized conifolds, we also discuss the relations between the quiver Yangians and some other Yangian algebras (and $\mathcal{W}$-algebras) in literature.
In this note, we study possible $\mathcal{R}$-matrix constructions in the context of quiver Yangians and Yang-Baxter algebras. For generalized conifolds, we also discuss the relations between the quiver Yangians and some other Yangian algebras (and $\mathcal{W}$-algebras) in literature.
△ Less
Submitted 10 August, 2022; v1 submitted 13 June, 2022;
originally announced June 2022.
-
Entire solutions of the generalized Hessian inequality
Authors:
Xiang Li,
Jing Hao,
Jiguang Bao
Abstract:
In this paper, we discuss the more general Hessian inequality $σ_{k}^{\frac{1}{k}}(λ(D_i (A\left(|Du|\right) D_j u)))\geq f(u)$ including the Laplacian, p-Laplacian, mean curvature, Hessian, k-mean curvature operators, and provide a necessary and sufficient condition on the global solvability, which can be regarded as generalized Keller-Osserman conditions.
In this paper, we discuss the more general Hessian inequality $σ_{k}^{\frac{1}{k}}(λ(D_i (A\left(|Du|\right) D_j u)))\geq f(u)$ including the Laplacian, p-Laplacian, mean curvature, Hessian, k-mean curvature operators, and provide a necessary and sufficient condition on the global solvability, which can be regarded as generalized Keller-Osserman conditions.
△ Less
Submitted 17 May, 2022;
originally announced May 2022.
-
A Contraction-constrained Model Predictive Control for Multi-timescale Nonlinear Processes
Authors:
Ryan McCloy,
Lai Wei,
Jie Bao
Abstract:
Many chemical processes exhibit diverse timescale dynamics with a strong coupling between timescale sensitive variables. Model predictive control with a non-uniformly spaced optimisation horizon is an effective approach to multi-timescale control and offers opportunities for reduced computational complexity. In such an approach the fast, moderate and slow dynamics can be included in the optimisati…
▽ More
Many chemical processes exhibit diverse timescale dynamics with a strong coupling between timescale sensitive variables. Model predictive control with a non-uniformly spaced optimisation horizon is an effective approach to multi-timescale control and offers opportunities for reduced computational complexity. In such an approach the fast, moderate and slow dynamics can be included in the optimisation problem by implementing smaller time intervals earlier in the prediction horizon and increasingly larger intervals towards the end of the prediction. In this paper, a reference-flexible condition is developed based on the contraction theory to provide a stability guarantee for a nonlinear system under non-uniform prediction horizons.
△ Less
Submitted 9 May, 2022;
originally announced May 2022.
-
A Contraction-constrained Model Predictive Control for Nonlinear Processes using Disturbance Forecasts
Authors:
Ryan McCloy,
Lai Wei,
Jie Bao
Abstract:
Model predictive control (MPC) has become the most widely used advanced control method in process industry. In many cases, forecasts of the disturbances are available, e.g., predicted renewable power generation based on weather forecast. While the predictions of disturbances may not be accurate, utilizing the information can significantly improve the control performance in response to the disturba…
▽ More
Model predictive control (MPC) has become the most widely used advanced control method in process industry. In many cases, forecasts of the disturbances are available, e.g., predicted renewable power generation based on weather forecast. While the predictions of disturbances may not be accurate, utilizing the information can significantly improve the control performance in response to the disturbances. By exploiting process and disturbance models, future system behaviour can be predicted and used to optimise control actions via minimisation of an economical cost function which incorporates these predictions. However, stability guarantee of the resulting closed-loop system is often difficult in this approach when the processes are nonlinear. Proposed in the following article is a contraction-constrained predictive controller which optimises process economy whilst ensuring stabilisation to operating targets subject to disturbance measurements and forecasts.
△ Less
Submitted 6 June, 2022; v1 submitted 9 May, 2022;
originally announced May 2022.
-
High-dimensional Asymptotics of Feature Learning: How One Gradient Step Improves the Representation
Authors:
Jimmy Ba,
Murat A. Erdogdu,
Taiji Suzuki,
Zhichao Wang,
Denny Wu,
Greg Yang
Abstract:
We study the first gradient descent step on the first-layer parameters $\boldsymbol{W}$ in a two-layer neural network: $f(\boldsymbol{x}) = \frac{1}{\sqrt{N}}\boldsymbol{a}^\topσ(\boldsymbol{W}^\top\boldsymbol{x})$, where $\boldsymbol{W}\in\mathbb{R}^{d\times N}, \boldsymbol{a}\in\mathbb{R}^{N}$ are randomly initialized, and the training objective is the empirical MSE loss:…
▽ More
We study the first gradient descent step on the first-layer parameters $\boldsymbol{W}$ in a two-layer neural network: $f(\boldsymbol{x}) = \frac{1}{\sqrt{N}}\boldsymbol{a}^\topσ(\boldsymbol{W}^\top\boldsymbol{x})$, where $\boldsymbol{W}\in\mathbb{R}^{d\times N}, \boldsymbol{a}\in\mathbb{R}^{N}$ are randomly initialized, and the training objective is the empirical MSE loss: $\frac{1}{n}\sum_{i=1}^n (f(\boldsymbol{x}_i)-y_i)^2$. In the proportional asymptotic limit where $n,d,N\to\infty$ at the same rate, and an idealized student-teacher setting, we show that the first gradient update contains a rank-1 "spike", which results in an alignment between the first-layer weights and the linear component of the teacher model $f^*$. To characterize the impact of this alignment, we compute the prediction risk of ridge regression on the conjugate kernel after one gradient step on $\boldsymbol{W}$ with learning rate $η$, when $f^*$ is a single-index model. We consider two scalings of the first step learning rate $η$. For small $η$, we establish a Gaussian equivalence property for the trained feature map, and prove that the learned kernel improves upon the initial random features model, but cannot defeat the best linear model on the input. Whereas for sufficiently large $η$, we prove that for certain $f^*$, the same ridge estimator on trained features can go beyond this "linear regime" and outperform a wide range of random features and rotationally invariant kernels. Our results demonstrate that even one gradient step can lead to a considerable advantage over random features, and highlight the role of learning rate scaling in the initial phase of training.
△ Less
Submitted 3 May, 2022;
originally announced May 2022.
-
Machine Learning Algebraic Geometry for Physics
Authors:
Jiakang Bao,
Yang-Hui He,
Elli Heyes,
Edward Hirst
Abstract:
We review some recent applications of machine learning to algebraic geometry and physics. Since problems in algebraic geometry can typically be reformulated as mappings between tensors, this makes them particularly amenable to supervised learning. Additionally, unsupervised methods can provide insight into the structure of such geometrical data. At the heart of this programme is the question of ho…
▽ More
We review some recent applications of machine learning to algebraic geometry and physics. Since problems in algebraic geometry can typically be reformulated as mappings between tensors, this makes them particularly amenable to supervised learning. Additionally, unsupervised methods can provide insight into the structure of such geometrical data. At the heart of this programme is the question of how geometry can be machine learned, and indeed how AI helps one to do mathematics. This is a chapter contribution to the book Machine learning and Algebraic Geometry, edited by A. Kasprzyk et al.
△ Less
Submitted 21 April, 2022;
originally announced April 2022.
-
Exponential ergodicity for damping Hamiltonian dynamics with state-dependent and non-local collisions
Authors:
Jianhai Bao,
Jian Wang
Abstract:
In this paper, we investigate the exponential ergodicity in a Wasserstein-type distance for a damping Hamiltonian dynamics with state-dependent and non-local collisions, which indeed is a special case of piecewise deterministic Markov processes while is very popular in numerous modelling situations including stochastic algorithms. The approach adopted in this work is based on a combination of the…
▽ More
In this paper, we investigate the exponential ergodicity in a Wasserstein-type distance for a damping Hamiltonian dynamics with state-dependent and non-local collisions, which indeed is a special case of piecewise deterministic Markov processes while is very popular in numerous modelling situations including stochastic algorithms. The approach adopted in this work is based on a combination of the refined basic coupling and the refined reflection coupling for non-local operators. In a certain sense, the main result developed in the present paper is a continuation of the counterpart in \cite{BW2022} on exponential ergodicity of stochastic Hamiltonian systems with Lévy noises and a complement of \cite{BA} upon exponential ergodicity for Andersen dynamics with constant jump rate functions.
△ Less
Submitted 4 April, 2022;
originally announced April 2022.
-
Liouville property and existence of entire solutions of Hessian equations
Authors:
Cong Wang,
Jiguang Bao
Abstract:
In this paper, we establish the existence and uniqueness theorem for entire solutions of Hessian equations with prescribed asymptotic behavior at infinity. This extends the previous results on Monge-Ampère equations. Our approach also makes the prescribed asymptotic order optimal within the range preset in exterior Dirichlet problems. In addition, we show a Liouville type result for $k$-convex sol…
▽ More
In this paper, we establish the existence and uniqueness theorem for entire solutions of Hessian equations with prescribed asymptotic behavior at infinity. This extends the previous results on Monge-Ampère equations. Our approach also makes the prescribed asymptotic order optimal within the range preset in exterior Dirichlet problems. In addition, we show a Liouville type result for $k$-convex solutions. This partly removes the $(k+1)$- or $n$-convexity restriction imposed in existing work.
△ Less
Submitted 4 March, 2022;
originally announced March 2022.
-
Amortized Proximal Optimization
Authors:
Juhan Bae,
Paul Vicol,
Jeff Z. HaoChen,
Roger Grosse
Abstract:
We propose a framework for online meta-optimization of parameters that govern optimization, called Amortized Proximal Optimization (APO). We first interpret various existing neural network optimizers as approximate stochastic proximal point methods which trade off the current-batch loss with proximity terms in both function space and weight space. The idea behind APO is to amortize the minimizatio…
▽ More
We propose a framework for online meta-optimization of parameters that govern optimization, called Amortized Proximal Optimization (APO). We first interpret various existing neural network optimizers as approximate stochastic proximal point methods which trade off the current-batch loss with proximity terms in both function space and weight space. The idea behind APO is to amortize the minimization of the proximal point objective by meta-learning the parameters of an update rule. We show how APO can be used to adapt a learning rate or a structured preconditioning matrix. Under appropriate assumptions, APO can recover existing optimizers such as natural gradient descent and KFAC. It enjoys low computational overhead and avoids expensive and numerically sensitive operations required by some second-order optimizers, such as matrix inverses. We empirically test APO for online adaptation of learning rates and structured preconditioning matrices for regression, image reconstruction, image classification, and natural language translation tasks. Empirically, the learning rate schedules found by APO generally outperform optimal fixed learning rates and are competitive with manually tuned decay schedules. Using APO to adapt a structured preconditioning matrix generally results in optimization performance competitive with second-order methods. Moreover, the absence of matrix inversion provides numerical stability, making it effective for low precision training.
△ Less
Submitted 28 February, 2022;
originally announced March 2022.
-
Crystal Melting, BPS Quivers and Plethystics
Authors:
Jiakang Bao,
Yang-Hui He,
Ali Zahabi
Abstract:
We study the refined and unrefined crystal/BPS partition functions of D6-D2-D0 brane bound states for all toric Calabi-Yau threefolds without compact 4-cycles and some non-toric examples. They can be written as products of (generalized) MacMahon functions. We check our expressions and use them as vacuum characters to study the gluings. We then consider the wall crossings and discuss possible cryst…
▽ More
We study the refined and unrefined crystal/BPS partition functions of D6-D2-D0 brane bound states for all toric Calabi-Yau threefolds without compact 4-cycles and some non-toric examples. They can be written as products of (generalized) MacMahon functions. We check our expressions and use them as vacuum characters to study the gluings. We then consider the wall crossings and discuss possible crystal descriptions for different chambers. We also express the partition functions in terms of plethystic exponentials. For $\mathbb{C}^3$ and tripled affine quivers, we find their connections to nilpotent Kac polynomials. Similarly, the partition functions of D4-D2-D0 brane bound states can be obtained by replacing the (generalized) MacMahon functions with the inverse of (generalized) Euler functions.
△ Less
Submitted 9 May, 2022; v1 submitted 25 February, 2022;
originally announced February 2022.
-
Random periodic solutions for stochastic differential equations with non-uniform dissipativity
Authors:
Jianhai Bao,
Yue Wu
Abstract:
This paper is concerned with the existence and uniqueness of random periodic solutions for stochastic differential equations (SDEs), where the drift terms involved need not to be uniformly dissipative. On the one hand, via the reflection coupling approach, we investigate the existence of random periodic solutions in the sense of distribution for SDEs without memory, where the drifts are merely dis…
▽ More
This paper is concerned with the existence and uniqueness of random periodic solutions for stochastic differential equations (SDEs), where the drift terms involved need not to be uniformly dissipative. On the one hand, via the reflection coupling approach, we investigate the existence of random periodic solutions in the sense of distribution for SDEs without memory, where the drifts are merely dissipative at long distance. On the other hand, via the synchronous coupling strategy, we establish respectively the existence of pathwise random periodic solutions for functional SDEs with a finite time lag and an infinite time lag, in which the drifts are only dissipative on average rather than uniformly dissipative with respect to the time parameters.
△ Less
Submitted 27 May, 2025; v1 submitted 20 February, 2022;
originally announced February 2022.
-
Asymptotic behavior of solutions to the Monge--Ampère equations with slow convergence rate at infinity
Authors:
Zixiao Liu,
Jiguang Bao
Abstract:
We consider the asymptotic behavior of solutions to the Monge--Ampère equations with slow convergence rate at infinity and fulfill previous results under faster convergence rate by Bao--Li--Zhang [Calc. Var PDE. 52(2015). pp. 39-63]. Different from known results, we obtain the limit of Hessian and/or gradient of solution at infinity relying on the convergence rate. The basic idea is to use a revis…
▽ More
We consider the asymptotic behavior of solutions to the Monge--Ampère equations with slow convergence rate at infinity and fulfill previous results under faster convergence rate by Bao--Li--Zhang [Calc. Var PDE. 52(2015). pp. 39-63]. Different from known results, we obtain the limit of Hessian and/or gradient of solution at infinity relying on the convergence rate. The basic idea is to use a revised level set method, the spherical harmonic expansion and the iteration method.
△ Less
Submitted 13 February, 2022;
originally announced February 2022.
-
Asymptotic expansion of 2-dimensional gradient graph with vanishing mean curvature at infinity
Authors:
Zixiao Liu,
Jiguang Bao
Abstract:
In this paper, we establish the asymptotic expansion at infinity of gradient graph in dimension 2 with vanishing mean curvature at infinity. This corresponds to our previous results in higher dimensions and generalizes the results for minimal gradient graph on exterior domain in dimension 2. Different from the strategies for higher dimensions, instead of the equivalence of Green's function on unbo…
▽ More
In this paper, we establish the asymptotic expansion at infinity of gradient graph in dimension 2 with vanishing mean curvature at infinity. This corresponds to our previous results in higher dimensions and generalizes the results for minimal gradient graph on exterior domain in dimension 2. Different from the strategies for higher dimensions, instead of the equivalence of Green's function on unbounded domains, we apply a version of iteration methods from Bao--Li--Zhang [Calc.Var PDE, 52(2015), pp. 39-63] that is refined by spherical harmonic expansions to provide a more explicit asymptotic behavior than known results.
△ Less
Submitted 10 February, 2022;
originally announced February 2022.
-
Quantum algorithm for stochastic optimal stopping problems with applications in finance
Authors:
João F. Doriguello,
Alessandro Luongo,
Jinge Bao,
Patrick Rebentrost,
Miklos Santha
Abstract:
The famous least squares Monte Carlo (LSM) algorithm combines linear least square regression with Monte Carlo simulation to approximately solve problems in stochastic optimal stopping theory. In this work, we propose a quantum LSM based on quantum access to a stochastic process, on quantum circuits for computing the optimal stopping times, and on quantum techniques for Monte Carlo. For this algori…
▽ More
The famous least squares Monte Carlo (LSM) algorithm combines linear least square regression with Monte Carlo simulation to approximately solve problems in stochastic optimal stopping theory. In this work, we propose a quantum LSM based on quantum access to a stochastic process, on quantum circuits for computing the optimal stopping times, and on quantum techniques for Monte Carlo. For this algorithm, we elucidate the intricate interplay of function approximation and quantum algorithms for Monte Carlo. Our algorithm achieves a nearly quadratic speedup in the runtime compared to the LSM algorithm under some mild assumptions. Specifically, our quantum algorithm can be applied to American option pricing and we analyze a case study for the common situation of Brownian motion and geometric Brownian motion processes.
△ Less
Submitted 27 July, 2023; v1 submitted 30 November, 2021;
originally announced November 2021.
