-
Periodic FPU system: Continuum limit to KdV via regularization and Fourier analysis
Authors:
Chulkwang Kwak,
Changhun Yang
Abstract:
The Fermi-Pasta-Ulam (FPU) system, initially introduced by Fermi for numerical simulations, models vibrating chains with fixed endpoints, where particles interact weakly, nonlinearly with their nearest neighbors. Contrary to the anticipated ergodic behavior, the simulation revealed nearly periodic (quasi-periodic) motion of the solutions, a phenomenon later referred to as the FPU paradox. A partia…
▽ More
The Fermi-Pasta-Ulam (FPU) system, initially introduced by Fermi for numerical simulations, models vibrating chains with fixed endpoints, where particles interact weakly, nonlinearly with their nearest neighbors. Contrary to the anticipated ergodic behavior, the simulation revealed nearly periodic (quasi-periodic) motion of the solutions, a phenomenon later referred to as the FPU paradox. A partial but remarkable explanation was provided by Zabusky and Kruskal [36], who formally derived the continuum limit of the FPU system, connecting it to the Korteweg-de Vries (KdV) equation. This formal derivation was later rigorously justified by Bambusi and Ponno [4].
In this paper, we revisit the problem studied in [4], specifically focusing on the continuum limit of the periodic FPU system for a broader class of initial data, as the number of particles N tends to infinity within a fixed domain. Unlike the non-periodic case discussed in [15], periodic FPU solutions lack a (local) smoothing effect, posing a significant challenge in controlling one derivative in the nonlinearity. This control is crucial not only for proving the (uniform in N) well-posedness for rough data but also for deriving the continuum limit. The main strategies to resolve this issue involve deriving L4-Strichartz estimates for FPU solutions, analogous to those previously derived for KdV solutions in [7], and regularizing the system via the normal form method introduced in [1].
△ Less
Submitted 2 February, 2025;
originally announced February 2025.
-
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
Authors:
Gemini Team,
Petko Georgiev,
Ving Ian Lei,
Ryan Burnell,
Libin Bai,
Anmol Gulati,
Garrett Tanzer,
Damien Vincent,
Zhufeng Pan,
Shibo Wang,
Soroosh Mariooryad,
Yifan Ding,
Xinyang Geng,
Fred Alcober,
Roy Frostig,
Mark Omernick,
Lexi Walker,
Cosmin Paduraru,
Christina Sorokin,
Andrea Tacchetti,
Colin Gaffney,
Samira Daruki,
Olcan Sercinoglu,
Zach Gleicher,
Juliette Love
, et al. (1112 additional authors not shown)
Abstract:
In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February…
▽ More
In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.
△ Less
Submitted 16 December, 2024; v1 submitted 8 March, 2024;
originally announced March 2024.
-
Gemini: A Family of Highly Capable Multimodal Models
Authors:
Gemini Team,
Rohan Anil,
Sebastian Borgeaud,
Jean-Baptiste Alayrac,
Jiahui Yu,
Radu Soricut,
Johan Schalkwyk,
Andrew M. Dai,
Anja Hauth,
Katie Millican,
David Silver,
Melvin Johnson,
Ioannis Antonoglou,
Julian Schrittwieser,
Amelia Glaese,
Jilin Chen,
Emily Pitler,
Timothy Lillicrap,
Angeliki Lazaridou,
Orhan Firat,
James Molloy,
Michael Isard,
Paul R. Barham,
Tom Hennigan,
Benjamin Lee
, et al. (1326 additional authors not shown)
Abstract:
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultr…
▽ More
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.
△ Less
Submitted 9 May, 2025; v1 submitted 18 December, 2023;
originally announced December 2023.
-
Energy solutions for the fifth-order modified Korteweg de-Vries equations
Authors:
Chulkwang Kwak,
Kiyeon Lee
Abstract:
We consider the Cauchy problem for the fifth-order modified Korteweg-de Vries equation (mKdV) under the periodic boundary condition. The fifth-order mKdV is an asymptotic model for shallow surface waves, and (in the perspective of integrable systems) the second equation in the mKdV hierarchy as well. In strong contrast with the non-periodic case, periodic solutions for dispersive equations do not…
▽ More
We consider the Cauchy problem for the fifth-order modified Korteweg-de Vries equation (mKdV) under the periodic boundary condition. The fifth-order mKdV is an asymptotic model for shallow surface waves, and (in the perspective of integrable systems) the second equation in the mKdV hierarchy as well. In strong contrast with the non-periodic case, periodic solutions for dispersive equations do not have a (local) smoothing effect, and this observation becomes a major obstacle to considering the Cauchy problem for dispersive equations under the periodic condition, consequently, the periodic fifth-order mKdV shows a quasilinear phenomenon, while the non-periodic case can be considered as a semilinear equation.
In this paper, we mainly establish the global well-posedness of the fifth-order mKdV in the energy space ($H^2(\mathbb T)$), which is an improvement of the former result by the first author (2018). The main idea to overcome the lack of (local) smoothing effect is to introduce a suitable (frequency dependent) short-time space originally motivated by the work by Ionescu, Kenig, and Tataru (2008). The new idea is to combine the (frequency) localized modified energy with additional weight in the spaces, which eventually handles the logarithmic divergence appearing in the energy estimates. Moreover, by using examples localized in low and very high frequencies, we show that the flow map of the fifth-order mKdV equation is not $C^3$, which implies that the Picard iterative method is not available for the local theory. This weakly concludes the quasilinear phenomenon of the periodic fifth-order mKdV.
△ Less
Submitted 12 May, 2023;
originally announced May 2023.
-
On the control issues for higher-order nonlinear dispersive equations on the circle
Authors:
Roberto de A. Capistrano-Filho,
Chulkwang Kwak,
Francisco J. Vielma Leal
Abstract:
The local and global control results for a general higher-order KdV-type operator posed on the unit circle are presented. Using spectral analysis, we are able to prove local results, that is, the equation is locally controllable and exponentially stable. To extend the local results to the global one we captured the smoothing properties of the Bourgain spaces, the so-called propagation of singulari…
▽ More
The local and global control results for a general higher-order KdV-type operator posed on the unit circle are presented. Using spectral analysis, we are able to prove local results, that is, the equation is locally controllable and exponentially stable. To extend the local results to the global one we captured the smoothing properties of the Bourgain spaces, the so-called propagation of singularities, which are proved with a new perspective. These propagation, together with the Strichartz estimates, are the key to extending the local control properties to the global one, precisely, higher-order KdV-type equations are globally controllable and exponentially stabilizable in the Sobolev space $H^{s}(\mathbb{T})$ for any $s \geq 0$. Our results recover previous results in the literature for the KdV and Kawahara equations and extend, for a general higher-order operator of KdV-type, the Strichartz estimates as well as the propagation results, which are the main novelties of this work.
△ Less
Submitted 29 September, 2021; v1 submitted 31 August, 2021;
originally announced September 2021.
-
Strichartz estimates for higher-order Schrödinger equations and their applications
Authors:
Younghun Hong,
Chulkwang Kwak,
Changhun Yang
Abstract:
In this paper, we consider the higher-order linear Schrödinger equations, that is, a formal finite Taylor expansion of the linear pseudo-relativistic equation. We establish the global-in-time Strichartz estimates for these higher-order equations which hold uniformly in the speed of light. As nonlinear applications, we show that the higher-order Hartree(-Fock) equation approximates the correspondin…
▽ More
In this paper, we consider the higher-order linear Schrödinger equations, that is, a formal finite Taylor expansion of the linear pseudo-relativistic equation. We establish the global-in-time Strichartz estimates for these higher-order equations which hold uniformly in the speed of light. As nonlinear applications, we show that the higher-order Hartree(-Fock) equation approximates the corresponding pseudo-relativistic equation on an arbitrarily long time interval, with higher accuracy than the non-relativistic equation. We also prove small data scattering for the higher-order nonlinear Schrödinger equations.
△ Less
Submitted 23 February, 2022; v1 submitted 25 June, 2021;
originally announced June 2021.
-
On the continuum limit for the discrete Nonlinear Schrödinger equation on a large finite cubic lattice
Authors:
Younghun Hong,
Chulkwang Kwak,
Changhun Yang
Abstract:
In this study, we consider the nonlinear Schödinger equation (NLS) with the zero-boundary condition on a two- or three-dimensional large finite cubic lattice. We prove that its solution converges to that of the NLS on the entire Euclidean space with simultaneous reduction in the lattice distance and expansion of the domain. Moreover, we obtain a precise global-in-time bound for the rate of converg…
▽ More
In this study, we consider the nonlinear Schödinger equation (NLS) with the zero-boundary condition on a two- or three-dimensional large finite cubic lattice. We prove that its solution converges to that of the NLS on the entire Euclidean space with simultaneous reduction in the lattice distance and expansion of the domain. Moreover, we obtain a precise global-in-time bound for the rate of convergence. Our proof heavily relies on Strichartz estimates on a finite lattice. A key observation is that, compared to the case of a lattice with a fixed size [Y. Hong, C. Kwak, S. Nakamura, and C. Yang, \emph{Finite difference scheme for two-dimensional periodic nonlinear {S}chrödinger equations}, Journal of Evolution Equations \textbf{21} (2021), no.~1, 391--418.], the loss of regularity in Strichartz estimates can be reduced as the domain expands, depending on the speed of expansion. This allows us to address the physically important three-dimensional case.
△ Less
Submitted 18 February, 2022; v1 submitted 25 June, 2021;
originally announced June 2021.
-
Characterization of Pd and Pd@Au core-shell nanoparticles using atom probe tomography and field evaporation simulation
Authors:
Se-Ho Kim,
Kyuseon Jang,
Phil Woong Kang,
Jae-Pyoung Ahn,
Jae-Bok Seol,
Chang-Min Kwak,
Constantinos Hatzoglou,
Francois Vurpillot,
Pyuck-Pa Choi
Abstract:
We report on atom probe tomography analyses of Pd and Pd@Au nanoparticles embedded in a Ni matrix and the effects of local evaporation field variations on the atom probe data. In order to assess the integrity of the reconstructed atom maps, we performed numerical simulations of the field evaporation processes and compared the simulated datasets with experimentally acquired data. The distortions se…
▽ More
We report on atom probe tomography analyses of Pd and Pd@Au nanoparticles embedded in a Ni matrix and the effects of local evaporation field variations on the atom probe data. In order to assess the integrity of the reconstructed atom maps, we performed numerical simulations of the field evaporation processes and compared the simulated datasets with experimentally acquired data. The distortions seen in the atom maps for both Pd and Pd@Au nanoparticles could be mostly ascribed to local variations in chemical composition and elemental evaporation fields. The evaporation field values for Pd and Ni, taken from the image hump model and assumed in the simulations, yielded a good agreement between experimental and simulation results. In contrast, the evaporation field for Au, as predicted from the image hump model, appeared to be substantially overestimated and resulted in a large discrepancy between experiments and simulations.
△ Less
Submitted 10 February, 2021;
originally announced February 2021.
-
Ill-posedness issues on $(abcd)$-Boussinesq system
Authors:
Chulkwang Kwak,
Christopher Maulén
Abstract:
In this paper, we consider the Cauchy problem for $(abcd)$-Boussinesq system posed on one- and two-dimensional Euclidean spaces. This model, initially introduced by Bona, Chen, and Saut, describes a small-amplitude waves on the surface of an inviscid fluid, and derived as a first-order approximation of incompressible, irrotational Euler equations. We mainly establish the ill-posedness of the syste…
▽ More
In this paper, we consider the Cauchy problem for $(abcd)$-Boussinesq system posed on one- and two-dimensional Euclidean spaces. This model, initially introduced by Bona, Chen, and Saut, describes a small-amplitude waves on the surface of an inviscid fluid, and derived as a first-order approximation of incompressible, irrotational Euler equations. We mainly establish the ill-posedness of the system under various parameter regimes, which generalize the result of the one-dimensional BBM-BBM case by Chen and Liu. Most of results established here, we obtain the optimal result for two-dimensional BBM-BBM system. The proof follows from an observation of the \emph{high to low-frequency cascade} present in nonlinearity, motivated by Bejenaru and Tao.
△ Less
Submitted 1 February, 2021;
originally announced February 2021.
-
Korteweg--de Vries limit for the Fermi--Pasta--Ulam system
Authors:
Younghun Hong,
Chulkwang Kwak,
Changhun Yang
Abstract:
In this paper, we develop dispersive PDE techniques for the Fermi--Pasta--Ulam (FPU) system with infinitely many oscillators, and we show that general solutions to the infinite FPU system can be approximated by counter-propagating waves governed by the Korteweg--de Vries (KdV) equation as the lattice spacing approaches zero. Our result not only simplifies the hypotheses but also reduces the regula…
▽ More
In this paper, we develop dispersive PDE techniques for the Fermi--Pasta--Ulam (FPU) system with infinitely many oscillators, and we show that general solutions to the infinite FPU system can be approximated by counter-propagating waves governed by the Korteweg--de Vries (KdV) equation as the lattice spacing approaches zero. Our result not only simplifies the hypotheses but also reduces the regularity requirement in the previous study [45].
△ Less
Submitted 27 December, 2019;
originally announced December 2019.
-
Finite difference scheme for two-dimensional periodic nonlinear Schrödinger equations
Authors:
Younghun Hong,
Chulkwang Kwak,
Shohei Nakamura,
Changhun Yang
Abstract:
A nonlinear Schrödinger equation (NLS) on a periodic box can be discretized as a discrete nonlinear Schrödinger equation (DNLS) on a periodic cubic lattice, which is a system of finitely many ordinary differential equations. We show that in two spatial dimensions, solutions to the DNLS converge strongly in $L^2$ to those of the NLS as the grid size $h>0$ approaches zero. As a result, the effective…
▽ More
A nonlinear Schrödinger equation (NLS) on a periodic box can be discretized as a discrete nonlinear Schrödinger equation (DNLS) on a periodic cubic lattice, which is a system of finitely many ordinary differential equations. We show that in two spatial dimensions, solutions to the DNLS converge strongly in $L^2$ to those of the NLS as the grid size $h>0$ approaches zero. As a result, the effectiveness of the finite difference method (FDM) is justified for the two-dimensional periodic NLS.
△ Less
Submitted 21 April, 2019;
originally announced April 2019.
-
Well-posedness issues on the periodic modified Kawahara equation
Authors:
Chulkwang Kwak
Abstract:
This paper is concerned with the Cauchy problem of the modified Kawahara equation (posed on $\mathbb T$), which is well-known as a model of capillary-gravity waves in an infinitely long canal over a flat bottom in a long wave regime \cite{Hasimoto1970}. We show in this paper some well-posedness results, mainly the \emph{global well-posedness} in $L^2(\mathbb T)$. The proof basically relies on the…
▽ More
This paper is concerned with the Cauchy problem of the modified Kawahara equation (posed on $\mathbb T$), which is well-known as a model of capillary-gravity waves in an infinitely long canal over a flat bottom in a long wave regime \cite{Hasimoto1970}. We show in this paper some well-posedness results, mainly the \emph{global well-posedness} in $L^2(\mathbb T)$. The proof basically relies on the idea introduced in Takaoka-Tsutsumi's works \cite{TT2004, NTT2010}, which weakens the non-trivial resonance in the cubic interactions (a kind of smoothing effect) for the local result, and the global well-posedness result immediately follows from $L^2$ conservation law. An immediate application of Takaoka-Tsutsumi's idea is available only in $H^s(\mathbb T)$, $s > 0$, due to the lack of $L^4$-Strichartz estimate for arbitrary $L^2$ data, a slight modification, thus, is needed to attain the local well-posedness in $L^2(\mathbb T)$. This is the first low regularity (global) well-posedness result for the periodic modified Kwahara equation, as far as we know. A direct interpolation argument ensures the \emph{unconditional uniqueness} in $H^s(\mathbb T)$, $s > \frac12$, and as a byproduct, we show the weak ill-posedness below $H^{\frac12}(\mathbb T)$, in the sense that the flow map fails to be uniformly continuous.
△ Less
Submitted 30 September, 2019; v1 submitted 24 February, 2019;
originally announced February 2019.
-
Asymptotic dynamics for the small data weakly dispersive one-dimensional Hamiltonian ABCD system
Authors:
Chulkwang Kwak,
Claudio Muñoz
Abstract:
Consider the Hamiltonian $abcd$ system in one dimension, with data posed in the energy space $H^1\times H^1$. This model, introduced by Bona, Chen and Saut, is a well-known physical generalization of the classical Boussinesq equations. The Hamiltonian case corresponds to the regime where $a,c<0$ and $b=d>0$. Under this regime, small solutions in the energy space are globally defined. A first proof…
▽ More
Consider the Hamiltonian $abcd$ system in one dimension, with data posed in the energy space $H^1\times H^1$. This model, introduced by Bona, Chen and Saut, is a well-known physical generalization of the classical Boussinesq equations. The Hamiltonian case corresponds to the regime where $a,c<0$ and $b=d>0$. Under this regime, small solutions in the energy space are globally defined. A first proof of decay for this $2\times 2$ system was given by the two authors and Poblete and Pozo, in a strongly dispersive regime, i.e. under essentially the conditions \[ b=d > \frac29, \quad a,c<-\frac1{18}. \] Additionally, decay was obtained inside a proper subset of the light cone $(-|t|,|t|)$. In this paper, we improve the last result in three directions. First, we enlarge the set of parameters $(a,b,c,d)$ for which decay to zero is the only available option, considering now the so-called weakly dispersive regime $a,c\sim 0$: we prove decay if now \[ b=d > \frac3{16}, \quad a,c<-\frac1{48}. \] This result is sharp in the case where $a=c$, since for $a,c$ bigger, some $abcd$ linear waves of nonzero frequency do have zero group velocity. Second, we sharply enlarge the interval of decay to consider the whole light cone, that is to say, any interval of the form $|x|\sim |v|t$, for any $|v|<1$. This result rules out, among other things, the existence of nonzero speed solitary waves in the regime where decay is present. Finally, we prove decay to zero of small $abcd$ solutions in exterior regions $|x|\gg |t|$, also discarding super-luminical small solitary waves. These three results are obtained by performing new improved virial estimates for which better decay properties are deduced.
△ Less
Submitted 22 April, 2019; v1 submitted 1 February, 2019;
originally announced February 2019.
-
Global solutions and stability properties of the 5th order Gardner equation
Authors:
Miguel A. Alejo,
Chulkwang Kwak
Abstract:
In this work, we deal with the initial value problem of the 5th-order Gardner equation in $\mathbb{R}$, presenting the local well-posedness result in $H^2(\mathbb{R})$. As a consequence of the local result, in addition to $H^2$-energy conservation law, we are able to prove the global well-posedness result in $H^2(\mathbb{R})$. Finally, we present a stability result for 5th order Gardner breather s…
▽ More
In this work, we deal with the initial value problem of the 5th-order Gardner equation in $\mathbb{R}$, presenting the local well-posedness result in $H^2(\mathbb{R})$. As a consequence of the local result, in addition to $H^2$-energy conservation law, we are able to prove the global well-posedness result in $H^2(\mathbb{R})$. Finally, we present a stability result for 5th order Gardner breather solution in the Sobolev space $H^2(\mathbb{R})$.
△ Less
Submitted 14 January, 2019; v1 submitted 10 January, 2019;
originally announced January 2019.
-
New Movement and Transformation Principle of Fuzzy Reasoning and Its Application to Fuzzy Neural Network
Authors:
Chung-Jin Kwak,
Son-Il Kwak,
Dae-Song Kang,
Song-Il Choe,
Jin-Ung Kim,
Hyok-Gi Chea
Abstract:
In this paper, we propose a new fuzzy reasoning principle, so called Movement and Transformation Principle(MTP). This Principle is to obtain a new fuzzy reasoning result by Movement and Transformation the consequent fuzzy set in response to the Movement, Transformation, and Movement-Transformation operations between the antecedent fuzzy set and fuzzificated observation information. And then we pre…
▽ More
In this paper, we propose a new fuzzy reasoning principle, so called Movement and Transformation Principle(MTP). This Principle is to obtain a new fuzzy reasoning result by Movement and Transformation the consequent fuzzy set in response to the Movement, Transformation, and Movement-Transformation operations between the antecedent fuzzy set and fuzzificated observation information. And then we presented fuzzy modus ponens and fuzzy modus tollens based on MTP. We compare proposed method with Mamdani fuzzy system, Sugeno fuzzy system, Wang distance type fuzzy reasoning method and Hellendoorn functional type method. And then we applied to the learning experiments of the fuzzy neural network based on MTP and compared it with the Sugeno method. Through prediction experiments of fuzzy neural network on the precipitation data and security situation data, learning accuracy and time performance are clearly improved. Consequently we show that our method based on MTP is computationally simple and does not involve nonlinear operations, so it is easy to handle mathematically.
△ Less
Submitted 9 November, 2018;
originally announced November 2018.
-
On the dynamics of zero-speed solutions for Camassa-Holm type equations
Authors:
Miguel A. Alejo,
Manuel F. Cortez,
Chulkwang Kwak,
Claudio Muñoz
Abstract:
In this paper we consider globally defined solutions of Camassa-Holm (CH) type equations outside the well-known nonzero speed, peakon region. These equations include the standard CH and Degasperis-Procesi (DP) equations, as well as nonintegrable generalizations such as the $b$-family, elastic rod and BBM equations. Having globally defined solutions for these models, we introduce the notion of \emp…
▽ More
In this paper we consider globally defined solutions of Camassa-Holm (CH) type equations outside the well-known nonzero speed, peakon region. These equations include the standard CH and Degasperis-Procesi (DP) equations, as well as nonintegrable generalizations such as the $b$-family, elastic rod and BBM equations. Having globally defined solutions for these models, we introduce the notion of \emph{zero-speed and breather solutions}, i.e., solutions that do not decay to zero as $t\to +\infty$ on compact intervals of space. We prove that, under suitable decay assumptions, such solutions do not exist because the identically zero solution is the global attractor of the dynamics, at least in a spatial interval of size $|x|\lesssim t^{1/2-}$ as $t\to+\infty$. As a consequence, we also show scattering and decay in CH type equations with long range nonlinearities. Our proof relies in the introduction of suitable Virial functionals à la Martel-Merle in the spirit of the works by one of us and Ponce, and Kowalczyk-Martel adapted to CH, DP and BBM type dynamics, one of them placed in $L^1_x$, and a second one in the energy space $H^1_x$. Both functionals combined lead to local in space decay to zero in $|x|\lesssim t^{1/2-}$ as $t\to+\infty$. Our methods do not rely on the integrable character of the equation, applying to other nonintegrable families of CH type equations as well.
△ Less
Submitted 22 October, 2018;
originally announced October 2018.
-
Local well-posedness of the fifth-order KdV-type equations on the half-line
Authors:
Márcio Cavalcante,
Chulkwang Kwak
Abstract:
This paper is a continuation of authors' previous work \cite{CK2018-1}. We extend the argument \cite{CK2018-1} to fifth-order KdV-type equations with different nonlinearities, in specific, where the scaling argument does not hold. We establish the $X^{s,b}$ nonlinear estimates for $b < \frac12$, which is almost optimal compared to the standard $X^{s,b}$ nonlinear estimates for $b > \frac12$ \cite{…
▽ More
This paper is a continuation of authors' previous work \cite{CK2018-1}. We extend the argument \cite{CK2018-1} to fifth-order KdV-type equations with different nonlinearities, in specific, where the scaling argument does not hold. We establish the $X^{s,b}$ nonlinear estimates for $b < \frac12$, which is almost optimal compared to the standard $X^{s,b}$ nonlinear estimates for $b > \frac12$ \cite{CGL2010, JH2009}. As an immediate conclusion, we prove the local well-posedness of the initial-boundary value problem (IBVP) for fifth-order KdV-type equations on the right half-line and the left half-line.
△ Less
Submitted 2 January, 2019; v1 submitted 16 August, 2018;
originally announced August 2018.
-
The initial-boundary value problem for the Kawahara equation on the half-line
Authors:
Márcio Cavalcante,
Chulkwang Kwak
Abstract:
This paper concerns the initial-boundary value problem (IBVP) of the Kawahara equation posed on the right and left half-lines. We prove the local well-posedness in the low regularity Sobolev space. We introduce the Duhamel boundary forcing operator, which is introduced by Colliander - Kenig \cite{CK} in the context of Airy group operators, to construct solutions on the whole line. We also give the…
▽ More
This paper concerns the initial-boundary value problem (IBVP) of the Kawahara equation posed on the right and left half-lines. We prove the local well-posedness in the low regularity Sobolev space. We introduce the Duhamel boundary forcing operator, which is introduced by Colliander - Kenig \cite{CK} in the context of Airy group operators, to construct solutions on the whole line. We also give the bilinear estimate in $X^{s,b}$ space for $b < \frac12$, which is almost sharp compared to IVP of Kawahara equation \cite{CLMW2009, JH2009}.
△ Less
Submitted 21 August, 2018; v1 submitted 14 May, 2018;
originally announced May 2018.
-
Extended decay properties for generalized BBM equations
Authors:
Chulkwang Kwak,
Claudio Muñoz
Abstract:
In this note we show that all small solutions of the BBM equation must decay to zero as $t\to +\infty$ in large portions of the physical space, extending previous known results, and only assuming data in the energy space. Our results also include decay on the left portion of the physical line, unlike the standard KdV dynamics.
In this note we show that all small solutions of the BBM equation must decay to zero as $t\to +\infty$ in large portions of the physical space, extending previous known results, and only assuming data in the energy space. Our results also include decay on the left portion of the physical line, unlike the standard KdV dynamics.
△ Less
Submitted 7 September, 2018; v1 submitted 6 February, 2018;
originally announced February 2018.
-
The scattering problem for the $abcd$ Boussinesq system in the energy space
Authors:
Chulkwang Kwak,
Claudio Muñoz,
Felipe Poblete,
Juan C. Pozo
Abstract:
The Boussinesq $abcd$ system is a 4-parameter set of equations posed in $\mathbb{R}_t \times \mathbb{R}_x$, originally derived by Bona, Chen and Saut as first order 2-wave approximations of the incompressible and irrotational, two dimensional water wave equations in the shallow water wave regime, in the spirit of the original Boussinesq derivation. Among many particular regimes, depending each of…
▽ More
The Boussinesq $abcd$ system is a 4-parameter set of equations posed in $\mathbb{R}_t \times \mathbb{R}_x$, originally derived by Bona, Chen and Saut as first order 2-wave approximations of the incompressible and irrotational, two dimensional water wave equations in the shallow water wave regime, in the spirit of the original Boussinesq derivation. Among many particular regimes, depending each of them in terms of the value of the parameters $(a,b,c,d)$ present in the equations, the "generic" regime is characterized by the setting $b,d>0$ and $a,c<0$. The system is hamiltonian if also $b=d$. The equations in this regime are globally well-posed in the energy space $H^1\times H^1$, provided one works with small solutions. In this paper, we investigate decay and the scattering problem in this regime, which is characterized as having (quadratic) long-range nonlinearities, very weak linear decay $O(t^{-1/3})$ because of the one dimensional setting, and existence of non scattering solutions (solitary waves). We prove, among other results, that for a sufficiently dispersive $abcd$ systems (characterized only in terms of parameters $a, b$ and $c$), all small solutions must decay to zero, locally strongly in the energy space, in proper subset of the light cone $|x|\leq |t|$. We prove this result by constructing three suitable virial functionals in the spirit of works by Kowalczyk, Martel and the second author, and more recently by the last three authors (valid for the simpler scalar "good Boussinesq" model), leading to global in time decay and control of all local $H^1\times H^1$ terms. No parity nor extra decay assumptions are needed to prove decay, only small solutions in the energy space.
△ Less
Submitted 28 December, 2017; v1 submitted 26 December, 2017;
originally announced December 2017.
-
Consideration on Example 2 of "An Algorithm of General Fuzzy InferenceWith The Reductive Property"
Authors:
Son-Il Kwak,
Oh-Chol Gwon,
Chung-Jin Kwak
Abstract:
In this paper, we will show that (1) the results about the fuzzy reasoning algoritm obtained in the paper "Computer Sciences Vol. 34, No.4, pp.145-148, 2007" according to the paper "IEEE Transactions On systems, Man and cybernetics, 18, pp.1049-1056, 1988" are correct; (2) example 2 in the paper "An Algorithm of General Fuzzy Inference With The Reductive Property" presented by He Ying-Si, Quan Hai…
▽ More
In this paper, we will show that (1) the results about the fuzzy reasoning algoritm obtained in the paper "Computer Sciences Vol. 34, No.4, pp.145-148, 2007" according to the paper "IEEE Transactions On systems, Man and cybernetics, 18, pp.1049-1056, 1988" are correct; (2) example 2 in the paper "An Algorithm of General Fuzzy Inference With The Reductive Property" presented by He Ying-Si, Quan Hai-Jin and Deng Hui-Wen according to the paper "An approximate analogical reasoning approach based on similarity measures" presented by Tursken I.B. and Zhong zhao is incorrect; (3) the mistakes in their paper are modified and then a calculation example of FMT is supplemented.
△ Less
Submitted 12 December, 2017;
originally announced December 2017.
-
Periodic fourth-order cubic NLS: Local well-posedness and Non-squeezing property
Authors:
Chulkwang Kwak
Abstract:
In this paper, we consider the cubic fourth-order nonlinear Schrödinger equation (4NLS) under the periodic boundary condition. We prove two results. One is the local well-posedness in $H^s$ with $-1/3 \le s < 0$ for the Cauchy problem of the Wick ordered 4NLS. The other one is the non-squeezing property for the flow map of 4NLS in the symplectic phase space $L^2(\mathbb{T})$. To prove the former w…
▽ More
In this paper, we consider the cubic fourth-order nonlinear Schrödinger equation (4NLS) under the periodic boundary condition. We prove two results. One is the local well-posedness in $H^s$ with $-1/3 \le s < 0$ for the Cauchy problem of the Wick ordered 4NLS. The other one is the non-squeezing property for the flow map of 4NLS in the symplectic phase space $L^2(\mathbb{T})$. To prove the former we used the ideas introduced in [Takaoka and Tsutsumi 2004] and [Nakanish et al 2010], and to prove the latter we used the ideas in [Colliander et al 2005].
△ Less
Submitted 23 January, 2018; v1 submitted 31 July, 2017;
originally announced August 2017.
-
An Approach to Controller Design Based on the Generalized Cloud Model
Authors:
UnSun Pak,
YongChol Sin,
ChungJin Kwak,
GyongIl Ryang
Abstract:
In this paper, an approach to controller design based on the cloud models, without using the analog plant model is presented.
In this paper, an approach to controller design based on the cloud models, without using the analog plant model is presented.
△ Less
Submitted 10 February, 2017;
originally announced February 2017.
-
Relative Distribution of Water Clusters at Temperature (300-3000K) and Pressure (1-500MPa)
Authors:
Yong-U Ri,
Chang-Sob Kwak,
Kye-Ryong Sin
Abstract:
At 300-3000K and 1-500MPa, variations of relative contents for small water clusters (H2O)n (n=1~6) were calculated by using statistical mechanical methods. First, 9 kinds of small water clusters were selected and their structures were optimized by using ab initio method. In the wide range of temperature (300-3000K) and pressure (1-500MPa), their equilibrium constants of reactions for formation of…
▽ More
At 300-3000K and 1-500MPa, variations of relative contents for small water clusters (H2O)n (n=1~6) were calculated by using statistical mechanical methods. First, 9 kinds of small water clusters were selected and their structures were optimized by using ab initio method. In the wide range of temperature (300-3000K) and pressure (1-500MPa), their equilibrium constants of reactions for formation of 9 kinds of water clusters were determined by using molecular partition function. Next, changes of contents (molar fractions) as function of temperature and pressure were estimated. The obtained results for small water clusters can be used to interpret temperature-pressure dependency of the average number for the hydrogen bonds in water clusters and redistribution of the water clusters at the ultrasonic cavitation reactions.
△ Less
Submitted 22 April, 2016;
originally announced April 2016.
-
Global well-posedness and Nonsqueezing property for the higher-order KdV-type flow
Authors:
Sunghyun Hong,
Chulkwang Kwak
Abstract:
In this paper, we prove that the periodic higher-order KdV-type equation \[\left\{\begin{array}{ll} \partial_t u + (-1)^{j+1} \partial_x^{2j+1}u + \frac12 \partial_x(u^2)=0, \hspace{1em} &(t,x) \in \mathbb{R} \times \mathbb{T}, \\ u(0,x) = u_0(x), &u_0 \in H^s(\mathbb{T}). \end{array} \right.\] is globally well-posed in $H^s$ for $s \ge -\frac{j}{2}$, $j \ge 3$. The proof of the global well-posedn…
▽ More
In this paper, we prove that the periodic higher-order KdV-type equation \[\left\{\begin{array}{ll} \partial_t u + (-1)^{j+1} \partial_x^{2j+1}u + \frac12 \partial_x(u^2)=0, \hspace{1em} &(t,x) \in \mathbb{R} \times \mathbb{T}, \\ u(0,x) = u_0(x), &u_0 \in H^s(\mathbb{T}). \end{array} \right.\] is globally well-posed in $H^s$ for $s \ge -\frac{j}{2}$, $j \ge 3$. The proof of the global well-posedness is based on "I-method" introduced by Colliander et al. \cite{CKSTT1}. To apply "I-method", we factorize the resonant functions by using the different ways from Hirayama \cite{Hirayama}. Furthermore, we prove the nonsqueezing property of the periodic higher-order KdV-type equation as well. The proof relies on Gromov's nonsqueezing theorem for the finite dimensional Hamiltonian system and approximation for the solution flow. More precisely, after taking the frequency truncation to the solution flow, we applied the nonsqueezing theorem and then the result is transferred to the infinite dimensional original flow. This argument was introduced by Kuksin \cite{Kuksin:1995ue} and made concretely by Bourgain \cite{Bourgain:1994tr} for 1D cubic NLS flow, and Colliander et. al. \cite{CKSTT3} for the KdV flow. One of our observation is that the higher-order KdV-type equation has the better modulation effect from the non-resonant interaction than KdV equation. Hence, unlike the work of Colliander et. al. \cite{CKSTT3}, we can get the nonsqueezing property for the solution flow without the Miura transform.
△ Less
Submitted 8 April, 2016; v1 submitted 12 November, 2015;
originally announced November 2015.
-
Local well-posedness for the fifth-order KdV equations on $\mathbb{T}$
Authors:
Chulkwang Kwak
Abstract:
This paper is a continuation of the paper \emph{Low regularity Cauchy problem for the fifth-order modified KdV equations on $\mathbb{T}$}. In this paper, we consider the fifth-order equation in the Korteweg-de Vries (KdV) hierarchy as following: \begin{equation*} \begin{cases} \partial_t u - \partial_x^5 u + 30u^2\partial_x u + 20 u\partial_x u \partial_x^3u + 10u \partial_x^3 u = 0, \hspace{1em}…
▽ More
This paper is a continuation of the paper \emph{Low regularity Cauchy problem for the fifth-order modified KdV equations on $\mathbb{T}$}. In this paper, we consider the fifth-order equation in the Korteweg-de Vries (KdV) hierarchy as following: \begin{equation*} \begin{cases} \partial_t u - \partial_x^5 u + 30u^2\partial_x u + 20 u\partial_x u \partial_x^3u + 10u \partial_x^3 u = 0, \hspace{1em} (t,x) \in \mathbb{R} \times \mathbb{T}, u(0,x) = u_0(x) \in H^s(\mathbb{T}) \end{cases}. \end{equation*}
We prove the local well-posedness of the fifth-order KdV equation for low regularity Sobolev initial data via the energy method. This paper follows almost same idea and argument as in the paper \emph{Low regularity Cauchy problem for the fifth-order modified KdV equations on $\mathbb{T}$}. Precisely, we use some conservation laws of the KdV Hamiltonians to observe the direction which the nonlinear solution evolves to. Besides, it is essential to use the short time $X^{s,b}$ spaces to control the nonlinear terms due to \emph{high $\times$ low $\Rightarrow$ high} interaction component in the non-resonant nonlinear term. We also use the localized version of the modified energy in order to obtain the energy estimate.
As an immediate result from a conservation law in the scaling sub-critical problem, we have the global well-posedness result in the energy space $H^2$.
△ Less
Submitted 11 February, 2016; v1 submitted 4 October, 2015;
originally announced October 2015.
-
Low regularity Cauchy problem for the fifth-order modified KdV equations on $\mathbb{T}$
Authors:
Chulkwang Kwak
Abstract:
In this paper, we consider the fifth-order modified Korteweg-de Vries (modified KdV) equation under the periodic boundary condition. We prove the local well-posedness in $H^s(\mathbb T)$, $s > 2$, via the energy method. The main tool is the short-time Fourier restriction norm method, which was first introduced in its current form by Ionescu, Kenig and Tataru [Global well-posedness of the KP-I init…
▽ More
In this paper, we consider the fifth-order modified Korteweg-de Vries (modified KdV) equation under the periodic boundary condition. We prove the local well-posedness in $H^s(\mathbb T)$, $s > 2$, via the energy method. The main tool is the short-time Fourier restriction norm method, which was first introduced in its current form by Ionescu, Kenig and Tataru [Global well-posedness of the KP-I initial-value problem in the energy space, Invent. Math. 173 (2) (2008) 265--304]. Besides, we use the frequency localized modified energy to control the high-low interaction component in the energy estimate. We remark that under the periodic setting, the integrable structure is very useful (but not necessary) to remove harmful terms in the nonlinearity and this work is the first low regularity well-posedness result for the fifth-order modified KdV equation.
△ Less
Submitted 15 May, 2018; v1 submitted 1 October, 2015;
originally announced October 2015.
-
Rough solutions of the fifth-order KdV equations
Authors:
Zihua Guo,
Chulkwang Kwak,
Soonsik Kwon
Abstract:
We consider the Cauchy problem of the fifth-order equation arising from the Korteweg-de Vries (KdV) hierarchy
u_t + u_{xxxxx} + c_1u_{x} u_{xx} + c_2u u_{x} = 0 x,t \in \R
We prove a priori bound of solutions for H^s(\R) with s >= 5/4 and the local well-posedness for s >= 2. The method is a short time X^{s,b} space, which is first developed by Ionescu-Kenig-Tataru in the context of the KP-I equa…
▽ More
We consider the Cauchy problem of the fifth-order equation arising from the Korteweg-de Vries (KdV) hierarchy
u_t + u_{xxxxx} + c_1u_{x} u_{xx} + c_2u u_{x} = 0 x,t \in \R
We prove a priori bound of solutions for H^s(\R) with s >= 5/4 and the local well-posedness for s >= 2. The method is a short time X^{s,b} space, which is first developed by Ionescu-Kenig-Tataru in the context of the KP-I equation. In addition, we use a weight on localized X^{s,b} structures to reduce the contribution of high-low frequency interaction where the low frequency has large modulation. As an immediate result from a conservation law, we have the fifth-order equation in the KdV hierarchy is globally well-posed in the energy space H^2.
△ Less
Submitted 19 June, 2012; v1 submitted 3 May, 2012;
originally announced May 2012.
-
Monte Carlo Simulation of Sinusoidally Modulated Superlattice Growth
Authors:
H. Jeong,
B. Kahng,
S. Lee,
C. Y. Kwak,
A. -L. Barabasi,
J. K. Furdyna
Abstract:
The fabrication of ZnSe/ZnTe superlattices grown by the process of rotating the substrate in the presence of an inhomogeneous flux distribution instead of successively closing and opening of source shutters is studied via Monte Carlo simulations. It is found that the concentration of each compound is sinusoidally modulated along the growth direction, caused by the uneven arrival of Se and Te ato…
▽ More
The fabrication of ZnSe/ZnTe superlattices grown by the process of rotating the substrate in the presence of an inhomogeneous flux distribution instead of successively closing and opening of source shutters is studied via Monte Carlo simulations. It is found that the concentration of each compound is sinusoidally modulated along the growth direction, caused by the uneven arrival of Se and Te atoms at a given point of the sample, and by the variation of the Te/Se ratio at that point due to the rotation of the substrate. In this way we obtain a ZnSe$_{1-x}$Te$_x$ alloy in which the composition $x$ varies sinusoidally along the growth direction. The period of the modulation is directly controlled by the rate of the substrate rotation. The amplitude of the compositional modulation is monotonous for small angular velocities of the substrate rotation, but is itself modulated for large angular velocities. The average amplitude of the modulation pattern decreases as the angular velocity of substrate rotation increases and the measurement position approaches the center of rotation. The simulation results are in good agreement with previously published experimental measurements on superlattices fabricated in this manner.
△ Less
Submitted 10 August, 2001;
originally announced August 2001.
