arXiv subjects

Hoa Nguyen

Publications and source records attributed to Hoa Nguyen.

16 recordsLinked to original sources

arXiv2026

Robustness in Sequential Decision Making under Evolving Uncertainty: Evidence from High-Frequency Market Making

We study sequential decision making under evolving uncertainty in high-frequency financial markets, where changing market dynamics continually challenge static decision policies. We show that robustness has two economically meaningful dimensions: uncertainty tolerance, which determines how much uncertainty the decision maker allows, and action robustness, which governs how conservatively decisions respond. Robustness is not merely protection against model misspecification, but a state-dependent mechanism that reshapes sequential decision behaviors. Simulation and empirical evidence show that action robustness has a substantially larger impact than uncertainty tolerance. Moreover, excessive robustness may reduce profitability in illiquid markets by limiting execution opportunities.

q-fin.TR↗

arXiv2026

CXL-ClusterSim: Modeling CXL-based Disaggregated Memory Cluster for Pooling and Sharing using gem5 and SST

Large-scale AI training and inference require hundreds of gigabytes to terabytes of DRAM with high peak to average utilization ratios, resulting in overprovisioning. In cloud computing, DRAM constitutes a significant share of the cost. Yet, as shown by recent articles, DRAM is heavily under utilized. Memory disaggregation is a solution to both these problems. With the advent of the CXL protocol, there is renewed interest in designing and optimizing computing systems with disaggregated memory. However, at present, there are limited simulation tools available for exploring the design space and evaluating the performance tradeoffs in computer systems with disaggregated memory. In this paper, we propose CXL-ClusterSim, a full-system modeling and simulation framework by combining the gem5 simulator for fidelity, with the Structural Simulation Toolkit (SST) for parallel simulation. We outline the challenges in creating this simulation infrastructure and present a design that is scalable, flexible, and reasonably fast to help computer architects to explore the design space of CXL-based disaggregated memory and identify new opportunities for hardware/software codesign and performance optimization.

cs.AR↗

arXiv2025

Pickle: Precise, Flexible Cross-Core Last-level Cache Data Prefetching for Irregular Memory Accesses

Graph analytics and sparse scientific workloads are dominated by parallel chains of data-dependent, long-latency memory accesses whose patterns are difficult for hardware to infer yet straightforward to express in software. Conventional hardware prefetchers attempt to recover this information from address streams alone, but false positives lead to substantial memory traffic overhead. Software-assisted approaches offer greater flexibility but still consume core limited resources. We propose Pickle, a software-defined, hardware-managed lastlevel cache (LLC) prefetcher that follows the decoupled access/execute philosophy. Pickle serves as an independent access engine, fully decoupled from core resources, that executes prefetch kernels sliced from the original application to bring data into the shared LLC ahead of demand. We evaluate Pickle using full-system, cycle-level simulation of a cluster of 8 high-performance cores, running all GAP benchmark suite algorithms across nine real-world graphs and irregular-access dominated scientific applications from the NAS parallel benchmark suite. Over a no-prefetching baseline, Pickle achieves 1.49x geomean speedup with only 2% DRAM traffic overhead on graph algorithms, and 1.53x with a 4.5% memory traffic reduction on NAS scatter/gather kernels. For reference, the state-of-the-art coreprivate indirect prefetcher achieves 1.40x but incurs 43% DRAM traffic overhead on graph workloads, and 1.36x at zero traffic overhead on scatter/gather kernels, illustrating the challenge of inferring irregular access patterns without application-level context. Pickle also composes transparently with private cache prefetchers: combining it with the state-of-the-art indirect or a simple stride prefetcher yields 1.65x-1.66x and 1.72x-1.84x geomean speedup on graph and NAS scatter/gather workloads, respectively.

cs.AR↗

arXiv2025

Choreographer: A Full-System Framework for Fine-Grained Tasks in Cache Hierarchies

In this paper, we introduce Choreographer, a simulation framework that enables a holistic system-level evaluation of fine-grained accelerators designed for latency-sensitive tasks. Unlike existing frameworks, Choreographer captures all hardware and software overheads in core-accelerator and cache-accelerator interactions, integrating a detailed gem5-based hardware stack featuring an AMBA coherent hub interface (CHI) mesh network and a complete Linux-based software stack. To facilitate rapid prototyping, it offers a C++ application programming interface and modular configuration options. Our detailed cache model provides accurate insights into performance variations caused by cache configurations, which are not captured by other frameworks. The framework is demonstrated through two case studies: a data-aware prefetcher for graph analytics workloads, and a quicksort accelerator. Our evaluation shows that the prefetcher achieves speedups between 1.08x and 1.88x by reducing memory access latency, while the quicksort accelerator delivers more than 2x speedup with minimal address translation overhead. These findings underscore the ability of Choreographer to model complex hardware-software interactions and optimize performance in small task offloading scenarios.

cs.AR↗

arXiv2025

QUT: A Unit Testing Framework for Quantum Subroutines

We present the architectural design and prototype implementation of QUT (Quantum Unit Testing), a framework for unit testing of quantum subroutines. The framework is developed with a focus on usability and simplicity, making the complex theoretical concepts behind quantum unit testing accessible to a wide range of users with diverse backgrounds. This is achieved through the implementation of polymorphic probabilistic assertions, whose evaluation methods adapt to the data types of the arguments used in assertion statements, which may vary according to the context-dependent semantics of quantum subroutines. These arguments can be represented as qubit measurement outcomes, density matrices, or Choi matrices. For each type, the architecture integrates a specific testing protocol - such as quantum process tomography, quantum state tomography, or Pearson's chi-squared test - while remaining flexible enough to incorporate additional protocols in the future. The framework is built on the Qiskit software stack, providing compatibility with a broad range of quantum hardware backends and simulation platforms. Drawing on the reasoning provided by the denotational semantics of quantum subroutines, this work also highlights the key distinctions between quantum unit testing and its classical counterpart.

quant-ph↗

arXiv2024

Using theory and experiments of spheres moving near boundaries to optimize the method of images for regularized Stokeslets

The general system of images for regularized Stokeslets (GSIRS) developed by Cortez and Varela (2015) is used extensively to model Stokes flow phenomena such as microorganisms swimming near a boundary. Our collaborative team uses dynamically similar scaled macroscopic experiments to test theories for forces and torques on spheres moving near a boundary and use these data and the method of regularized Stokeslets (MRS) created by Cortez et al. (2015) to calibrate the GSIRS. We find excellent agreement between theory and experiments, which provides the first experimental validation of such series solutions for spheres moving near an infinite plane boundary. We test two surface discretization methods commonly used in the literature: the six-patch method and the spherical centroidal Voronoi tessellation (SCVT) method. Our data show that the SCVT method provides the most accurate results when the motional symmetry is broken by the presence of a boundary. We use theory and the MRS to find optimal values for the regularization parameter in free space and show that the optimal regularization parameter values can be fit with simple formulae when using the SCVT method, so that other researchers have an easy reference. We also present a regularization function with higher order accuracy when compared with the regularization function previously introduced by Cortez et al. (2005). The simulated force and torque values compare very well with experiments and theory for a wide range of boundary distances. But, we find the simulations lose accuracy when the gap between the edge of the sphere and the wall is smaller than the average distance between grid points using the SCVT method. Our computational parameters and MATLAB and PYTHON implementations of the Lee and Leal (1980) theory provide researchers with important resources to optimize the numerical simulations of spheres moving near boundaries.

physics.flu-dyn↗

arXiv2024

Is there really a Citation Age Bias in NLP?

Citations are a key ingredient of scientific research to relate a paper to others published in the community. Recently, it has been noted that there is a citation age bias in the Natural Language Processing (NLP) community, one of the currently fastest growing AI subfields, in that the mean age of the bibliography of NLP papers has become ever younger in the last few years, leading to `citation amnesia' in which older knowledge is increasingly forgotten. In this work, we put such claims into perspective by analyzing the bibliography of $\sim$300k papers across 15 different scientific fields submitted to the popular preprint server Arxiv in the time period from 2013 to 2022. We find that all AI subfields (in particular: cs.AI, cs.CL, cs.CV, cs.LG) have similar trends of citation amnesia, in which the age of the bibliography has roughly halved in the last 10 years (from above 12 in 2013 to below 7 in 2022), on average. Rather than diagnosing this as a citation age bias in the NLP community, we believe this pattern is an artefact of the dynamics of these research fields, in which new knowledge is produced in ever shorter time intervals.

cs.DL↗

arXiv2023

Revisiting LARS for Large Batch Training Generalization of Neural Networks

This paper explores Large Batch Training techniques using layer-wise adaptive scaling ratio (LARS) across diverse settings, uncovering insights. LARS algorithms with warm-up tend to be trapped in sharp minimizers early on due to redundant ratio scaling. Additionally, a fixed steep decline in the latter phase restricts deep neural networks from effectively navigating early-phase sharp minimizers. Building on these findings, we propose Time Varying LARS (TVLARS), a novel algorithm that replaces warm-up with a configurable sigmoid-like function for robust training in the initial phase. TVLARS promotes gradient exploration early on, surpassing sharp optimizers and gradually transitioning to LARS for robustness in later phases. Extensive experiments demonstrate that TVLARS consistently outperforms LARS and LAMB in most cases, with up to 2\% improvement in classification scenarios. Notably, in all self-supervised learning cases, TVLARS dominates LARS and LAMB with performance improvements of up to 10\%.

cs.LG↗

arXiv2022

BMX: Boosting Natural Language Generation Metrics with Explainability

State-of-the-art natural language generation evaluation metrics are based on black-box language models. Hence, recent works consider their explainability with the goals of better understandability for humans and better metric analysis, including failure cases. In contrast, our proposed method BMX: Boosting Natural Language Generation Metrics with explainability explicitly leverages explanations to boost the metrics' performance. In particular, we perceive feature importance explanations as word-level scores, which we convert, via power means, into a segment-level score. We then combine this segment-level score with the original metric to obtain a better metric. Our tests show improvements for multiple metrics across MT and summarization datasets. While improvements in machine translation are small, they are strong for summarization. Notably, BMX with the LIME explainer and preselected parameters achieves an average improvement of 0.087 points in Spearman correlation on the system-level evaluation of SummEval.

cs.CL↗

arXiv2022

Hydra: Hybrid Server Power Model

With the growing complexity of big data workloads that require abundant data and computation, data centers consume a tremendous amount of power daily. In an effort to minimize data center power consumption, several studies developed power models that can be used for job scheduling either reducing the number of active servers or balancing workloads across servers at their peak energy efficiency points. Due to increasing software and hardware heterogeneity, we observed that there is no single power model that works the best for all server conditions. Some complicated machine learning models themselves incur performance and power overheads and hence it is not desirable to use them frequently. There are no power models that consider containerized workload execution. In this paper, we propose a hybrid server power model, Hydra, that considers both prediction accuracy and performance overhead. Hydra dynamically chooses the best power model for the given server conditions. Compared with state-of-the-art solutions, Hydra outperforms across all compute-intensity levels on heterogeneous servers.

cs.DC↗

arXiv2021

Using Experimentally Calibrated Regularized Stokeslets to Assess Bacterial Flagellar Motility Near a Surface

The presence of a nearby boundary is likely to be important in the life cycle and evolution of motile flagellate bacteria. This has led many authors to employ numerical simulations to model near-surface bacterial motion and compute hydrodynamic boundary effects. A common choice has been the method of images for regularized Stokeslets (MIRS); however, the method requires discretization sizes and regularization parameters that are not specified by any theory. To determine appropriate regularization parameters for given discretization choices in MIRS, we conducted dynamically similar macroscopic experiments and fit the simulations to the data. In the experiments, we measured the torque on cylinders and helices of different wavelengths as they rotated in a viscous fluid at various distances to a boundary. We found that differences between experiments and optimized simulations were less than 5\% when using surface discretizations for cylinders and centerline discretizations for helices. Having determined optimal regularization parameters, we used MIRS to simulate an idealized free-swimming bacterium constructed of a cylindrical cell body and a helical flagellum moving near a boundary. We assessed the swimming performance of many bacterial morphologies by computing swimming speed, motor rotation rate, Purcell's propulsive efficiency, energy cost per distance, and a new metabolic energy cost defined to be the energy cost per body mass per distance. All five measures predicted the same optimal flagellar wavelength independently of body size and surface proximity. Although the measures disagreed on the optimal body size, they all predicted that body size is an important factor in the energy cost of bacterial motility near and far from a surface.

physics.bio-ph↗

arXiv2020

The gem5 Simulator: Version 20.0+

The open-source and community-supported gem5 simulator is one of the most popular tools for computer architecture research. This simulation infrastructure allows researchers to model modern computer hardware at the cycle level, and it has enough fidelity to boot unmodified Linux-based operating systems and run full applications for multiple architectures including x86, Arm, and RISC-V. The gem5 simulator has been under active development over the last nine years since the original gem5 release. In this time, there have been over 7500 commits to the codebase from over 250 unique contributors which have improved the simulator by adding new features, fixing bugs, and increasing the code quality. In this paper, we give and overview of gem5's usage and features, describe the current state of the gem5 simulator, and enumerate the major changes since the initial release of gem5. We also discuss how the gem5 simulator has transitioned to a formal governance model to enable continued improvement and community support for the next 20 years of computer architecture research.

cs.AR↗

arXiv2020

Block Model Guided Unsupervised Feature Selection

Feature selection is a core area of data mining with a recent innovation of graph-driven unsupervised feature selection for linked data. In this setting we have a dataset $\mathbf{Y}$ consisting of $n$ instances each with $m$ features and a corresponding $n$ node graph (whose adjacency matrix is $\mathbf{A}$) with an edge indicating that the two instances are similar. Existing efforts for unsupervised feature selection on attributed networks have explored either directly regenerating the links by solving for $f$ such that $f(\mathbf{y}_i,\mathbf{y}_j) \approx \mathbf{A}_{i,j}$ or finding community structure in $\mathbf{A}$ and using the features in $\mathbf{Y}$ to predict these communities. However, graph-driven unsupervised feature selection remains an understudied area with respect to exploring more complex guidance. Here we take the novel approach of first building a block model on the graph and then using the block model for feature selection. That is, we discover $\mathbf{F}\mathbf{M}\mathbf{F}^T \approx \mathbf{A}$ and then find a subset of features $\mathcal{S}$ that induces another graph to preserve both $\mathbf{F}$ and $\mathbf{M}$. We call our approach Block Model Guided Unsupervised Feature Selection (BMGUFS). Experimental results show that our method outperforms the state of the art on several real-world public datasets in finding high-quality features for clustering.

cs.LG↗

arXiv2020

Enhanced computational performance of the lattice Boltzmann model for simulating micron- and submicron-size particle flows and non-Newtonian fluid flows

Significant improvements in the computational performance of the lattice-Boltzmann (LB) model, coded in FORTRAN90, were achieved through application of enhancement techniques. Applied techniques include optimization of array memory layouts, data structure simplification, random number generation outside the simulation thread(s), code parallelization via OpenMP, and intra- and inter-timestep task pipelining. Effectiveness of these optimization techniques was measured on three benchmark problems: (i) transient flow of multiple particles in a Newtonian fluid in a heterogeneous fractured porous domain, (ii) thermal fluctuation of the fluid at the sub-micron scale and the resultant Brownian motion of a particle, and (iii) non-Newtonian fluid flow in a smooth-walled channel. Application of the aforementioned optimization techniques resulted in an average 21 performance improvement, which could significantly enhance practical uses of the LB models in diverse applications, focusing on the fate and transport of nano-size or micron-size particles in non-Newtonian fluids.

physics.comp-ph↗

arXiv2011

Multilingual Schema Matching for Wikipedia Infoboxes

Recent research has taken advantage of Wikipedia's multilingualism as a resource for cross-language information retrieval and machine translation, as well as proposed techniques for enriching its cross-language structure. The availability of documents in multiple languages also opens up new opportunities for querying structured Wikipedia content, and in particular, to enable answers that straddle different languages. As a step towards supporting such queries, in this paper, we propose a method for identifying mappings between attributes from infoboxes that come from pages in different languages. Our approach finds mappings in a completely automated fashion. Because it does not require training data, it is scalable: not only can it be used to find mappings between many language pairs, but it is also effective for languages that are under-represented and lack sufficient training samples. Another important benefit of our approach is that it does not depend on syntactic similarity between attribute names, and thus, it can be applied to language pairs that have distinct morphologies. We have performed an extensive experimental evaluation using a corpus consisting of pages in Portuguese, Vietnamese, and English. The results show that not only does our approach obtain high precision and recall, but it also outperforms state-of-the-art techniques. We also present a case study which demonstrates that the multilingual mappings we derive lead to substantial improvements in answer quality and coverage for structured queries over Wikipedia content.

cs.DB↗

arXiv2011

Synthesizing Products for Online Catalogs

A high-quality, comprehensive product catalog is essential to the success of Product Search engines and shopping sites such as Yahoo! Shopping, Google Product Search or Bing Shopping. But keeping catalogs up-to-date becomes a challenging task, calling for the need of automated techniques. In this paper, we introduce the problem of product synthesis, a key component of catalog creation and maintenance. Given a set of offers advertised by merchants, the goal is to identify new products and add them to the catalog together with their (structured) attributes. A fundamental challenge is the scale of the problem: a Product Search engine receives data from thousands of merchants and millions of products; the product taxonomy contains thousands of categories, where each category comes in a different schema; and merchants use representations for products that are different from the ones used in the catalog of the Product Search engine. We propose a system that provides an end-to-end solution to the product synthesis problem, and includes components for extraction, and addresses issues involved in data extraction from offers, schema reconciliation, and data fusion. We developed a novel and scalable technique for schema matching which leverages knowledge about previously-known instance-level associations between offers and products; and it is trained using automatically created training sets (no manually-labeled data is needed). We present an experimental evaluation of our system using data from Bing Shopping for more than 800K offers, a thousand merchants, and 400 categories. The evaluation confirms that our approach is able to automatically generate a large number of accurate product specifications, and that our schema reconciliation component outperforms state-of-the-art schema matching techniques in terms of precision and recall.

cs.DB↗