arXiv subjects

Ziyu Huang

Publications and source records attributed to Ziyu Huang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Surrogate Modeling of Landau Damping with Deep Operator Networks

Kinetic simulations excel at capturing microscale plasma physics phenomena with high accuracy, but their computational demands make them impractical for modeling large-scale space and astrophysical systems. In this context, we build a surrogate model, using Deep Operator Networks (DeepONets), based upon the Vlasov-Poisson simulation data to model the dynamical evolution of plasmas, focusing on the Landau damping process - a fundamental kinetic phenomenon in space and astrophysical plasmas. The trained DeepONets are able to capture the evolution of electric field energy in both linear and nonlinear regimes under various conditions. Extensive validation highlights DeepONets' robust performance in reproducing complex plasma behaviors with high accuracy, paving the way for large-scale modeling of space and astrophysical plasmas.

physics.plasm-ph↗

arXiv2025

First JWST thermal phase curves of temperate terrestrial exoplanets reveal no thick atmosphere around TRAPPIST-1 b and c

We report JWST/MIRI 15 $μ$m phase curves of TRAPPIST-1 b and c, revealing thermal emission consistent with their irradiation levels, assuming no efficient heat redistribution. We find that TRAPPIST-1 b shows a high dayside brightness temperature (490 $\pm$ 17 K), no significantly detectable nightside emission ($F_{\rm b, Night, max}$ = $39_{-27}^{+55}$ ppm), and no phase offset -- features consistent with a low-albedo, airless ultramafic rocky surface. TRAPPIST-1 c exhibits a lower dayside brightness temperature (369 $\pm$ 23 K), and a nightside flux statistically indistinguishable from that of TRAPPIST-1 b ($F_{\rm c, Night, max}$ = $62_{-43}^{+60}$ ppm). Atmosphere models with surface pressures $\geq$1 bar and efficient greenhouse effects are strongly disfavoured for both planets. TRAPPIST-1 b is unlikely to possess any substantial atmosphere, while TRAPPIST-1 c may retain a tenuous, greenhouse-poor O$_2$-dominated atmosphere or be similarly airless with a more reflective surface. These results suggest divergent evolutionary pathways or atmospheric loss processes, despite similar compositions. These measurements tightly constrain atmosphere retention in the inner TRAPPIST-1 system.

astro-ph.EP↗

arXiv2025

ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive

Large language model (LLM) decoding suffers from high latency due to fragmented execution across operators and heavy reliance on off-chip memory for data exchange and reduction. This execution model limits opportunities for fusion and incurs significant memory traffic and kernel launch overhead. While modern architectures such as NVIDIA Hopper provide distributed shared memory and low-latency intra-cluster interconnects, they expose only low-level data movement instructions, lacking structured abstractions for collective on-chip communication. To bridge this software-hardware gap, we introduce two cluster-level communication primitives, ClusterReduce and ClusterGather, which abstract common communication patterns and enable structured, high-speed data exchange and reduction between thread blocks within a cluster, allowing intermediate results to be on-chip without involving off-chip memory. Building on these abstractions, we design ClusterFusion, an execution framework that schedules communication and computation jointly to expand operator fusion scope by composing decoding stages such as QKV Projection, Attention, and Output Projection into a single fused kernels. Evaluations on H100 GPUs show that ClusterFusion outperforms state-of-the-art inference frameworks by 1.61x on average in end-to-end latency across different models and configurations. The source code is available at https://github.com/xinhao-luo/ClusterFusion.

cs.DC↗

arXiv2025

Exoplanet Atmospheric Escape Observations with the Habitable Worlds Observatory

The Decadal Survey on Astronomy and Astrophysics 2020 highlights the importance of advancing research focused on discovering and characterizing habitable worlds. In line with this priority, our goal is to investigate how planetary systems evolve through atmospheric escape and to develop methods for identifying potentially Earth-like planets. By leveraging the ultraviolet (UV) capabilities of the Habitable Worlds Observatory (HWO), we can use transit spectroscopy to observe atmospheric escape in exoplanets and explore the processes that shape their evolution, assess the ability of small planets to retain their atmospheres, and search for signs of Earth-like atmospheres. To achieve this, we support the development of a UV spectrograph with moderate- to high-resolution capabilities for point-source observations, coverage of key spectral features in the 100-300 nm range, and detectors that can register high count rates reliably. This article is an adaptation of a science case document developed for the Characterizing Exoplanets Steering Committee within HWO's Solar Systems in Context Working Group.

astro-ph.IM↗

arXiv2025

Deeply Optimizing the SAT Solver for the IC3 Algorithm

The IC3 algorithm, also known as PDR, is a SAT-based model checking algorithm that has significantly influenced the field in recent years due to its efficiency, scalability, and completeness. It utilizes SAT solvers to solve a series of SAT queries associated with relative induction. In this paper, we introduce several optimizations for the SAT solver in IC3 based on our observations of the unique characteristics of these SAT queries. By observing that SAT queries do not necessarily require decisions on all variables, we compute a subset of variables that need to be decided before each solving process while ensuring that the result remains unaffected. Additionally, noting that the overhead of binary heap operations in VSIDS is non-negligible, we replace the binary heap with buckets to achieve constant-time operations. Furthermore, we support temporary clauses without the need to allocate a new activation variable for each solving process, thereby eliminating the need to reset solvers. We developed a novel lightweight CDCL SAT solver, GipSAT, which integrates these optimizations. A comprehensive evaluation highlights the performance improvements achieved by GipSAT. Specifically, the GipSAT-based IC3 demonstrates an average speedup of 3.61 times in solving time compared to the IC3 implementation based on MiniSat.

cs.LO↗

arXiv2025

The rIC3 Hardware Model Checker

In this paper, we present rIC3, an efficient bit-level hardware model checker primarily based on the IC3 algorithm. It boasts a highly efficient implementation and integrates several recently proposed optimizations, such as the specifically optimized SAT solver, dynamically adjustment of generalization strategies, and the use of predicates with internal signals, among others. As a first-time participant in the Hardware Model Checking Competition, rIC3 was independently evaluated as the best-performing tool, not only in the bit-level track but also in the word-level bit-vector track through bit-blasting. Our experiments further demonstrate significant advancements in both efficiency and scalability. rIC3 can also serve as a backend for verifying industrial RTL designs using SymbiYosys. Additionally, the source code of rIC3 is highly modular, with the IC3 algorithm module being particularly concise, making it an academic platform that is easy to modify and extend.

cs.FL↗

arXiv2025

Second Order Fully Nonlinear Mean Field Games with Degenerate Diffusions

In this article, we study the global-in-time well-posedness of second order mean field games (MFGs) with both nonlinear drift functions simultaneously depending on the state, distribution and control variables, and the diffusion term depending on both state and distribution. Besides, the diffusion term is allowed to be degenerate, unbounded and even nonlinear in the distribution, but it does not depend on the control. First, we establish the global well-posedness of the corresponding forward-backward stochastic differential equations (FBSDEs), which arise from the maximum principle under a so-called $β$-monotonicity commonly used in the optimal control theory. The $β$-monotonicity admits more interesting cases, as representative examples including but not limited to the displacement monotonicity, the small mean field effect condition or the Lasry-Lions monotonicity; and ensures the well-posedness result in diverse non-convex examples. In our settings, we pose assumptions directly on the drift and diffusion coefficients and the cost functionals, rather than indirectly on the Hamiltonian, to make the conditions more visible. Our probabilistic method tackles the nonlinear dynamics with a linear but infinite dimensional version, and together with our recently proposed cone property for the adjoint processes, following in an almost straightforward way the conventional approach to the classical stochastic control problem, we derive a sufficiently good regularity of the value functional, and finally show that it is the unique classical solution to the MFG master equation. Our results require fairly few conditions on the functional coefficients for solution of the MFG, and a bit more conditions -- which are least stringent in the contemporary literature -- for classical solution of the MFG master equation.

math.OC↗

arXiv2025

Machine-learning heat flux closure for multi-moment fluid modeling of nonlinear Landau damping

Nonlinear plasma physics problems are usually simulated through comprehensive modeling of phase space. The extreme computational cost of such simulations has motivated the development of multi-moment fluid models. However, a major challenge has been finding a suitable fluid closure for these fluid models. Recent developments in physics-informed machine learning have led to a renewed interest in constructing accurate fluid closure terms. In this study, we take an approach that integrates kinetic physics from the first-principles Vlasov simulations into a fluid model (through the heat flux closure term) using the Fourier neural operator - a neural network architecture. Without resolving the phase space dynamics, this new fluid model is capable of capturing the nonlinear evolution of the Landau damping process that exactly matches the Vlasov simulation results. This machine learning-assisted new approach provides a computationally affordable framework that surpasses previous fluid models in accurately modeling the kinetic evolution of complex plasma systems.

physics.plasm-ph↗

arXiv2025

Extended CTG Generalization and Dynamic Adjustment of Generalization Strategies in IC3

The IC3 algorithm is widely used in hardware formal verification, with generalization as a crucial step. Standard generalization expands a cube by dropping literals to include more unreachable states. The CTG approach enhances this by blocking counterexamples to generalization (CTG) when dropping literals fails. In this paper, we extend the CTG method (EXCTG) to put more effort into generalization. If blocking the CTG fails, EXCTG attempts to block its predecessors, aiming for better generalization. While CTG and EXCTG offer better generalization results, they also come with increased computational overhead. Finding an appropriate balance between generalization quality and computational overhead is challenging with a static strategy. We propose DynAMic, a method that dynamically adjusts generalization strategies according to the difficulty of blocking states, thereby improving scalability without compromising efficiency. A comprehensive evaluation demonstrates that EXCTG and DynAMic achieve significant scalability improvements, solving 8 and 25 more cases, respectively, compared to CTG generalization.

cs.FL↗

arXiv2024

On Mean Field Monotonicity Conditions from Control Theoretical Perspective

In this article, from the viewpoint of control theory, we discuss the relationships among the commonly used monotonicity conditions that ensure the well-posedness of the solutions arising from problems of mean field games (MFGs) and mean field type control (MFTC). We first introduce the well-posedness of general forward-backward stochastic differential equations (FBSDEs) defined on some suitably chosen Hilbert spaces under the $β$-monotonicity. We then propose a monotonicity condition for the MFG, namely partitioning the running cost functional into two parts, so that both parts still depend on the control and the state distribution, yet one satisfies a strong convexity and a small mean field effect condition, while the other has a newly introduced displacement quasi-monotonicity. To the best of our knowledge, the latter quasi type condition has not yet been discussed in the contemporary literature, and it can be considered as a bit more general monotonicity condition than those commonly used. Besides, for the MFG, we show that convexity and small mean field effect condition for the first part of running cost functional and the quasi-monotonicity condition for the second part together imply the $β$-monotonicity and thus the well-posedness for the associated FBSDEs. For the MFTC problem, we show that the $β$-monotonicity for the corresponding FBSDEs is simply the convexity assumption on the cost functional. Finally, we consider a more general setting where the drift functional is allowed to be non-linear for both MFG and MFTC problems.

math.OC↗

arXiv2024

Mean Field Games with Major and Minor Agents: the Limiting Problem and Nash Equilibrium

In this paper, we consider a mean field game (MFG) with a major and $N$ minor agents. We first consider the limiting problem and allow the coefficients to vary with the conditional distribution in a nonlinear way. We use the stochastic maximum principle to transform the limiting control problem into a system of two coupled conditional distribution dependent forward-backward stochastic differential equations (FBSDEs), and prove the existence and uniqueness result of the FBSDEs when the dependence between major agent and minor agents is sufficiently weak. We then use the solution of the limiting problem to construct an $\mathcal{O}(N^{-\frac{1}{2}})$-Nash equilibrium for the MFG with a major and $N$ minor agents.

math.OC↗

arXiv2024

Probing Cold-to-Temperate Exoplanetary Atmospheres: The Role of Water Condensation on Surface Identification with JWST

Understanding the surface temperature and interior structure of cold-to-temperate sub-Neptunes is critical for assessing their habitability, yet direct observations are challenging. In this study, we investigate the impact of water condensation on the atmospheric compositions of sub-Neptunes, focusing on the implications for JWST spectroscopic observations. By modeling the atmospheric photochemistry of two canonical sub-Neptunes, K2-18 b and LHS 1140 b, both with and without water condensation and with and without thick atmospheres, we demonstrate that water condensation can significantly affect the predicted atmospheric compositions. This effect is driven by oxygen depletion from the condensation of water vapor and primarily manifests as an increase in the C/O ratio within the photochemically active regions of the atmosphere. This change in composition particularly affects planets with thin H2-dominated atmospheres, leading to a transition in dominant nitrogen and carbon carriers from N2 and oxygen-rich species like CO/CO2 towards heavier hydrocarbons and nitriles. While our models do not fully account for the loss mechanisms of these higher-order species, such molecules can go on to form more refractory molecules or hazes. Planets with thin H2-rich atmospheres undergoing significant water condensation are thus likely to exhibit very hazy atmospheres. The relatively flat JWST spectra observed for LHS 1140 b could be consistent with such a scenario, suggesting a shallow surface with extensive water condensation or a high atmospheric C/O ratio. Conversely, the JWST observations of K2-18 b are better aligned with a volatile-rich mini-Neptune with a thick atmosphere.

astro-ph.EP↗

arXiv2024

A roadmap for the atmospheric characterization of terrestrial exoplanets with JWST

Ultra-cool dwarf stars are abundant, long-lived, and uniquely suited to enable the atmospheric study of transiting terrestrial companions with JWST. Amongst them, the most prominent is the M8.5V star TRAPPIST-1 and its seven planets. While JWST Cycle 1 observations have started to yield preliminary insights into the planets, they have also revealed that their atmospheric exploration requires a better understanding of their host star. Here, we propose a roadmap to characterize the TRAPPIST-1 system -- and others like it -- in an efficient and robust manner. We notably recommend that -- although more challenging to schedule -- multi-transit windows be prioritized to mitigate the effects of stellar activity and gather up to twice more transits per JWST hour spent. We conclude that, for such systems, planets cannot be studied in isolation by small programs, but rather need large-scale, jointly space- and ground-based initiatives to fully exploit the capabilities of JWST for the exploration of terrestrial planets.

astro-ph.EP↗

arXiv2024

Inexact and Implementable Accelerated Newton Proximal Extragradient Method for Convex Optimization

In this paper, we investigate the convergence behavior of the Accelerated Newton Proximal Extragradient (A-NPE) method when employing inexact Hessian information. The exact A-NPE method was the pioneer near-optimal second-order approach, exhibiting an oracle complexity of $\Tilde{O}(ε^{-2/7})$ for convex optimization. Despite its theoretical optimality, there has been insufficient attention given to the study of its inexact version and efficient implementation. We introduce the inexact A-NPE method (IA-NPE), which is shown to maintain the near-optimal oracle complexity. In particular, we design a dynamic approach to balance the computational cost of constructing the Hessian matrix and the progress of the convergence. Moreover, we show the robustness of the line-search procedure, which is a subroutine in IA-NPE, in the face of the inexactness of the Hessian. These nice properties enable the implementation of highly effective machine learning techniques like sub-sampling and various heuristics in the method. Extensive numerical results illustrate that IA-NPE compares favorably with state-of-the-art second-order methods, including Newton's method with cubic regularization and Trust-Region methods.

math.OC↗

arXiv2024

GMLake: Efficient and Transparent GPU Memory Defragmentation for Large-scale DNN Training with Virtual Memory Stitching

Large-scale deep neural networks (DNNs), such as large language models (LLMs), have revolutionized the artificial intelligence (AI) field and become increasingly popular. However, training or fine-tuning such models requires substantial computational power and resources, where the memory capacity of a single acceleration device like a GPU is one of the most important bottlenecks. Owing to the prohibitively large overhead (e.g., $10 \times$) of GPUs' native memory allocator, DNN frameworks like PyTorch and TensorFlow adopt a caching allocator that maintains a memory pool with a splitting mechanism for fast memory (de)allocation. Unfortunately, the caching allocator's efficiency degrades quickly for popular memory reduction techniques such as recomputation, offloading, distributed training, and low-rank adaptation. The primary reason is that those memory reduction techniques introduce frequent and irregular memory (de)allocation requests, leading to severe fragmentation problems for the splitting-based caching allocator. To mitigate this fragmentation problem, we propose a novel memory allocation framework based on low-level GPU virtual memory management called GPU memory lake (GMLake). GMLake employs a novel virtual memory stitching (VMS) mechanism, which can fuse or combine non-contiguous memory blocks with a virtual memory address mapping. GMLake can reduce an average of 9.2 GB (up to 25 GB) GPU memory usage and 15% (up to 33% ) fragmentation among eight LLM models on GPU A100 with 80 GB memory. GMLake is completely transparent to the DNN models and memory reduction techniques and ensures the seamless execution of resource-intensive deep-learning tasks. We have open-sourced GMLake at https://github.com/intelligent-machine-learning/glake/tree/main/GMLake.

cs.DC↗

arXiv2023

Degenerate Mean Field Type Control with Linear and Unbounded Diffusion, and their Associated Equations

We study the well-posedness of a system of forward-backward stochastic differential equations (FBSDEs) corresponding to a degenerate mean field type control problem, when the diffusion coefficient depends on the state together with its measure and also the control. Degenerate mean field type control problems are rarely studied in the literature. Our method is based on a lifting approach which embeds the control problem and the associated FBSDEs in Wasserstein spaces into certain Hilbert spaces. We use a continuation method to establish the solvability of the FBSDEs and that of the Gâteaux derivatives of this FBSDEs. We then explore the regularity of the value function in time and in measure argument, and we also show that it is the unique classical solution of the associated Bellman equation. We also study the higher regularity of the linear functional derivative of the value function, by then, we obtain the classical solution of the mean field type master equation.

math.PR↗

arXiv2023

Maximum Principle for Mean Field Type Control Problems with General Volatility Functions

In this paper, we study the maximum principle of mean field type control problems when the volatility function depends on the state and its measure and also the control, by using our recently developed method. Our method is to embed the mean field type control problem into a Hilbert space to bypass the evolution in the Wasserstein space. We here give a necessary condition and a sufficient condition for these control problems in Hilbert spaces, and we also derive a system of forward-backward stochastic differential equations.

math.OC↗

arXiv2022

Fully Coupled Nonlocal Quasilinear Forward-Backward Parabolic Equations Arising from Mean Field Games

In this paper, we study fully coupled nonlocal second order quasilinear forward-backward partial differential equations (FBPDEs), which arise from solution of the mean field game (MFG) suggested by Lasry and Lions [Japan. J. Math. 2 (2007), p. 237 (Remark iv)]. We show the existence of solutions $(u,m)\in C^{1+\frac{1}{4},2+\frac{1}{2}}([0,T]\times\mathbb{R}^n)\times C^{\frac{1}{2}}([0,T],\mathcal{P}_1(\mathbb{R}^n))$, and also the uniqueness under an additional monotonicity condition. Then, we improve the regularity of our weak solution $m$ to get a classical solution under appropriate regularity assumptions on coefficients. The FBPDEs can be used to investigate a system of mean field equations (MFEs), where the backward one is a Hamilton-Jacobi-Bellman equation and the forward one is a Fokker-Planck equation. Moreover, we prove a verification theorem and give an optimal strategy of the associated MFG via the solution of MFEs. Finally, we address the linear-quadratic problems.

math.PR↗