Hung Nguyen

Van Nguyen, Trung Le, Chakkrit Tantithamthavorn et al.

Software vulnerabilities are a serious and crucial concern. Typically, in a program or function consisting of hundreds or thousands of source code statements, there are only a few statements causing the corresponding vulnerabilities. Most current approaches to vulnerability labelling are done on a function or program level by experts with the assistance of machine learning tools. Extending this approach to the code statement level is much more costly and time-consuming and remains an open problem. In this paper, we propose a novel end-to-end deep learning-based approach to identify the vulnerability-relevant code statements of a specific function. Inspired by the specific structures observed in real-world vulnerable code, we first leverage mutual information for learning a set of latent variables representing the relevance of the source code statements to the corresponding function's vulnerability. We then propose novel clustered spatial contrastive learning in order to further improve the representation learning and the robust selection process of vulnerability-relevant code statements. Experimental results on real-world datasets of 200k+ C/C++ functions show the superiority of our method over other state-of-the-art baselines. In general, our method obtains a higher performance in VCP, VCA, and Top-10 ACC measures of between 3% to 14% over the baselines when running on real-world datasets in an unsupervised setting. Our released source code samples are publicly available at \href{https://github.com/vannguyennd/livuitcl}{https://github.com/vannguyennd/livuitcl.}

Van Nguyen, Trung Le, Chakkrit Tantithamthavorn et al.

Software vulnerabilities (SVs) have become a common, serious and crucial concern due to the ubiquity of computer software. Many machine learning-based approaches have been proposed to solve the software vulnerability detection (SVD) problem. However, there are still two open and significant issues for SVD in terms of i) learning automatic representations to improve the predictive performance of SVD, and ii) tackling the scarcity of labeled vulnerabilities datasets that conventionally need laborious labeling effort by experts. In this paper, we propose a novel end-to-end approach to tackle these two crucial issues. We first exploit the automatic representation learning with deep domain adaptation for software vulnerability detection. We then propose a novel cross-domain kernel classifier leveraging the max-margin principle to significantly improve the transfer learning process of software vulnerabilities from labeled projects into unlabeled ones. The experimental results on real-world software datasets show the superiority of our proposed method over state-of-the-art baselines. In short, our method obtains a higher performance on F1-measure, the most important measure in SVD, from 1.83% to 6.25% compared to the second highest method in the used datasets. Our released source code samples are publicly available at https://github.com/vannguyennd/dam2p

Krishnamurthy Dvijotham, Hung Nguyen, Konstantin Turitsyn

Quadratic systems of equations appear in several applications. The results in this paper are motivated by quadratic systems of equations that describe equilibrium behavior of physical infrastructure networks like the power and gas grids. The quadratic systems in infrastructure networks are parameterized- the parameters can represent uncertainty (estimation error in resistance/inductance of a power transmission line, for example)or controllable decision variables (power outputs of generators,for example). It is then of interest to understand conditions on the parameters under which the quadratic system is guaranteed to have a solution within a specified set (for example, bounds on voltages and flows in a power grid). Given nominal values of the parameters at which the quadratic system has a solution and the Jacobian of the quadratic system at the solution i snon-singular, we develop a general framework to construct convex regions around the nominal value such that the system is guaranteed to have a solution within a given distance of the nominal solution. We show that several results from recen tliterature can be recovered as special cases of our framework,and demonstrate our approach on several benchmark power systems.

Mingyu Guo, Jialiang Li, Aneta Neumann et al.

We propose a combinatorial optimisation model called Limited Query Graph Connectivity Test. We consider a graph whose edges have two possible states (On/Off). The edges' states are hidden initially. We could query an edge to reveal its state. Given a source s and a destination t, we aim to test s-t connectivity by identifying either a path (consisting of only On edges) or a cut (consisting of only Off edges). We are limited to B queries, after which we stop regardless of whether graph connectivity is established. We aim to design a query policy that minimizes the expected number of queries. Our model is mainly motivated by a cyber security use case where we need to establish whether an attack path exists in a network, between a source and a destination. Edge query is resolved by manual effort from the IT admin, which is the motivation behind query minimization. Our model is highly related to monotone Stochastic Boolean Function Evaluation (SBFE). There are two existing exact algorithms for SBFE that are prohibitively expensive. We propose a significantly more scalable exact algorithm. While previous exact algorithms only scale for trivial graphs (i.e., past works experimented on at most 20 edges), we empirically demonstrate that our algorithm is scalable for a wide range of much larger practical graphs (i.e., Windows domain network graphs with tens of thousands of edges). We propose three heuristics. Our best-performing heuristic is via reducing the search horizon of the exact algorithm. The other two are via reinforcement learning (RL) and Monte Carlo tree search (MCTS). We also derive an anytime algorithm for computing the performance lower bound. Experimentally, we show that all our heuristics are near optimal. The exact algorithm based heuristic outperforms all, surpassing RL, MCTS and 8 existing heuristics ported from SBFE and related literature.

G. Dumindu Samaraweera, Hung Nguyen, Hadi Zanddizari et al.

Recent breakthrough technological progressions of powerful mobile computing resources such as low-cost mobile GPUs along with cutting-edge, open-source software architectures have enabled high-performance deep learning on mobile platforms. These advancements have revolutionized the capabilities of today's mobile applications in different dimensions to perform data-driven intelligence locally, particularly for smart health applications. Unlike traditional machine learning (ML) architectures, modern on-device deep learning frameworks are proficient in utilizing computing resources in mobile platforms seamlessly, in terms of producing highly accurate results in less inference time. However, on the flip side, energy resources in a mobile device are typically limited. Hence, whenever a complex Deep Neural Network (DNN) architecture is fed into the on-device deep learning framework, while it achieves high prediction accuracy (and performance), it also urges huge energy demands during the runtime. Therefore, managing these resources efficiently within the spectrum of performance and energy efficiency is the newest challenge for any mobile application featuring data-driven intelligence beyond experimental evaluations. In this paper, first, we provide a timely review of recent advancements in on-device deep learning while empirically evaluating the performance metrics of current state-of-the-art ML architectures and conventional ML approaches with the emphasis given on energy characteristics by deploying them on a smart health application. With that, we are introducing a new framework through an energy-aware, adaptive model comprehension and realization (EAMCR) approach that can be utilized to make more robust and efficient inference decisions based on the available computing/energy resources in the mobile device during the runtime.

Jialin Yuan, Jay Patravali, Hung Nguyen et al.

Unsupervised Video Object Segmentation (UVOS) aims at discovering objects and tracking them through videos. For accurate UVOS, we observe if one can locate precise segment proposals on key frames, subsequent processes are much simpler. Hence, we propose to reason about key frame proposals using a graph built with the object probability masks initially generated from multiple frames around the key frame and then propagated to the key frame. On this graph, we compute maximal cliques, with each clique representing one candidate object. By making multiple proposals in the clique to vote for the key frame proposal, we obtain refined key frame proposals that could be better than any of the single-frame proposals. A semi-supervised VOS algorithm subsequently tracks these key frame proposals to the entire video. Our algorithm is modular and hence can be used with any instance segmentation and semi-supervised VOS algorithm. We achieve state-of-the-art performance on the DAVIS-2017 validation and test-dev dataset. On the related problem of video instance segmentation, our method shows competitive performance with the previous best algorithm that requires joint training with the VOS algorithm.

Trong-Thang Pham, Nhat Le, Tuong Do et al.

Audio-driven talking head animation is a challenging research topic with many real-world applications. Recent works have focused on creating photo-realistic 2D animation, while learning different talking or singing styles remains an open problem. In this paper, we present a new method to generate talking head animation with learnable style references. Given a set of style reference frames, our framework can reconstruct 2D talking head animation based on a single input image and an audio stream. Our method first produces facial landmarks motion from the audio stream and constructs the intermediate style patterns from the style reference images. We then feed both outputs into a style-aware image generator to generate the photo-realistic and fidelity 2D animation. In practice, our framework can extract the style information of a specific character and transfer it to any new static image for talking head animation. The intensive experimental results show that our method achieves better results than recent state-of-the-art approaches qualitatively and quantitatively.

Hung Nguyen, Hans Moen, Pekka Marttinen

Extracting insights from Electronic Health Record (EHR) databases often requires SQL expertise, creating a barrier for clinical decision-making and research. A promising approach is to use Large Language Models (LLMs) to translate natural language questions into SQL through Retrieval-Augmented Generation (RAG), where relevant question-SQL examples are retrieved to generate new queries via few-shot learning. However, adapting this method to the medical domain is non-trivial, as effective retrieval requires examples that align with both the logical structure of the question and its referenced entities (e.g., drug names, procedure titles). Standard single-step RAG struggles to optimize both aspects simultaneously and often relies on near-exact matches to generalize effectively. This issue is especially severe in healthcare, as questions often contain noisy and inconsistent medical jargon. To address this, we present CBR-to-SQL, a framework inspired by Case-based Reasoning theory that decomposes RAG's single-step retrieval into two explicit stages: one that focuses on retrieving structurally relevant examples, and one that aligns entities with the target database schema. Evaluated on two clinical benchmarks, CBR-to-SQL achieves competitive accuracies compared to fine-tuned methods. More importantly, it demonstrates considerably higher sample efficiency and robustness than the standard RAG approach, particularly under data scarcity and retrieval perturbations.

Gheorghe Comanici, Eric Bieber, Mike Schaekermann et al. · amazon-science, baidu

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.

Hai Nguyen, Hieu Dao, Hung Nguyen et al.

This study presents high-throughput, real-time multi-agent affective computing framework designed to enhance classroom learning through emotional state monitoring. As large classroom sizes and limited teacher student interaction increasingly challenge educators, there is a growing need for scalable, data-driven tools capable of capturing students' emotional and engagement patterns in real time. The system was evaluated using the Classroom Emotion Dataset, consisting of 1,500 labeled images and 300 classroom detection videos. Tailored for IoT devices, the system addresses load balancing and latency challenges through efficient real-time processing. Field testing was conducted across three educational institutions in a large metropolitan area: a primary school (hereafter school A), a secondary school (school B), and a high school (school C). The system demonstrated robust performance, detecting up to 50 faces at 25 FPS and achieving 88% overall accuracy in classifying classroom engagement states. Implementation results showed positive outcomes, with favorable feedback from students, teachers, and parents regarding improved classroom interaction and teaching adaptation. Key contributions of this research include establishing a practical, IoT-based framework for emotion-aware learning environments and introducing the 'Classroom Emotion Dataset' to facilitate further validation and research.

Hung Nguyen, Chanho Kim, Rigved Naukarkar et al.

Long-term point tracking is essential to understand non-rigid motion in the physical world better. Deep learning approaches have recently been incorporated into long-term point tracking, but most prior work predominantly functions in 2D. Although these methods benefit from the well-established backbones and matching frameworks, the motions they produce do not always make sense in the 3D physical world. In this paper, we propose the first deep learning framework for long-term point tracking in 3D that generalizes to new points and videos without requiring test-time fine-tuning. Our model contains a cost volume fusion module that effectively integrates multiple past appearances and motion information via a transformer architecture, significantly enhancing overall tracking performance. In terms of 3D tracking performance, our model significantly outperforms simple scene flow chaining and previous 2D point tracking methods, even if one uses ground truth depth and camera pose to backproject 2D point tracks in a synthetic scenario.

Nhat Nguyen, Duong Nguyen, Gianluca Rizzo et al.

Decentralized planning is a key element of cooperative multi-agent systems for information gathering tasks. However, despite the high frequency of agent failures in realistic large deployment scenarios, current approaches perform poorly in the presence of failures, by not converging at all, and/or by making very inefficient use of resources (e.g. energy). In this work, we propose Attritable MCTS (A-MCTS), a decentralized MCTS algorithm capable of timely and efficient adaptation to changes in the set of active agents. It is based on the use of a global reward function for the estimation of each agent's local contribution, and regret matching for coordination. We evaluate its effectiveness in realistic data-harvesting problems under different scenarios. We show both theoretically and experimentally that A-MCTS enables efficient adaptation even under high failure rates. Results suggest that, in the presence of frequent failures, our solution improves substantially over the best existing approaches in terms of global utility and scalability.

Huy Quang Ngo, Mingyu Guo, Hung Nguyen

We study a Stackelberg game between an attacker and a defender on large Active Directory (AD) attack graphs where the defender employs a set of honeypots to stop the attacker from reaching high-value targets. Contrary to existing works that focus on small and static attack graphs, AD graphs typically contain hundreds of thousands of nodes and edges and constantly change over time. We consider two types of attackers: a simple attacker who cannot observe honeypots and a competent attacker who can. To jointly solve the game, we propose a mixed-integer programming (MIP) formulation. We observed that the optimal blocking plan for static graphs performs poorly in dynamic graphs. To solve the dynamic graph problem, we re-design the mixed-integer programming formulation by combining m MIP (dyMIP(m)) instances to produce a near-optimal blocking plan. Furthermore, to handle a large number of dynamic graph instances, we use a clustering algorithm to efficiently find the m-most representative graph instances for a constant m (dyMIP(m)). We prove a lower bound on the optimal blocking strategy for dynamic graphs and show that our dyMIP(m) algorithms produce close to optimal results for a range of AD graphs under realistic conditions.

Hung Nguyen, Peiyuan Wu, Morris Chang

One of the most challenging issues in federated learning is that the data is often not independent and identically distributed (nonIID). Clients are expected to contribute the same type of data and drawn from one global distribution. However, data are often collected in different ways from different resources. Thus, the data distributions among clients might be different from the underlying global distribution. This creates a weight divergence issue and reduces federated learning performance. This work focuses on improving federated learning performance for skewed data distribution across clients. The main idea is to adjust the client distribution closer to the global distribution using sample weights. Thus, the machine learning model converges faster with higher accuracy. We start from the fundamental concept of empirical risk minimization and theoretically derive a solution for adjusting the distribution skewness using sample weights. To determine sample weights, we implicitly exchange density information by leveraging a neural network-based density estimation model, MADE. The clients data distribution can then be adjusted without exposing their raw data. Our experiment results on three real-world datasets show that the proposed method not only improves federated learning accuracy but also significantly reduces communication costs compared to the other experimental methods.

Nhat Nguyen, Duong Nguyen, Gianluca Rizzo et al.

Decentralized Monte Carlo Tree Search (Dec-MCTS) is widely used for cooperative multi-agent planning but struggles in sparse or skewed reward environments. We introduce Coordinated Boltzmann MCTS (CB-MCTS), which replaces deterministic UCT with a stochastic Boltzmann policy and a decaying entropy bonus for sustained yet focused exploration. While Boltzmann exploration has been studied in single-agent MCTS, applying it in multi-agent systems poses unique challenges. CB-MCTS is the first to address this. We analyze CB-MCTS in the simple-regret setting and show in simulations that it outperforms Dec-MCTS in deceptive scenarios and remains competitive on standard benchmarks, providing a robust solution for multi-agent planning.

Shengkun Cui, Archit Patke, Hung Nguyen et al.

This study characterizes GPU resilience in Delta HPC, a large-scale AI system that consists of 1,056 A100 and H100 GPUs, with over 1,300 petaflops of peak throughput. Delta HPC is operated by the National Center for Supercomputing Applications (NCSA) at the University of Illinois Urbana-Champaign. We used 2.5 years of operational data (11.7 million GPU hours) on GPU errors. Our major findings include: (i) H100 GPU memory resilience is worse than A100 GPU memory, with 3.2x lower per-GPU MTBE for memory errors, (ii) The GPU memory error-recovery mechanisms on H100 GPUs are insufficient to handle the increased memory capacity, (iii) H100 GPUs demonstrate significantly improved GPU hardware resilience over A100 GPUs with respect to critical hardware components, (iv) GPU errors on both A100 and H100 GPUs frequently result in job failures due to the lack of robust recovery mechanisms at the application level, and (v) We project the impact of GPU node availability on larger-scales and find that significant overprovisioning of 5% is necessary to handle GPU failures.

Hung Nguyen, Blark Runfa Li, Truong Nguyen

Neural Radiance Fields (NeRF) has achieved superior performance in novel view synthesis and 3D scene representation, but its practical applications are hindered by slow convergence and reliance on dense training views. To this end, we present DWTNeRF, a unified framework based on Instant-NGP's fast-training hash encoding. It is coupled with regularization terms designed for few-shot NeRF, which operates on sparse training views. Our DWTNeRF additionally includes a novel Discrete Wavelet loss that allows explicit prioritization of low frequencies directly in the training objective, reducing few-shot NeRF's overfitting on high frequencies in earlier training stages. We also introduce a model-based approach, based on multi-head attention, that is compatible with INGP, which are sensitive to architectural changes. On the 3-shot LLFF benchmark, DWTNeRF outperforms Vanilla INGP by 15.07% in PSNR, 24.45% in SSIM and 36.30% in LPIPS. Our approach encourages a re-thinking of current few-shot approaches for fast-converging implicit representations like INGP or 3DGS.

Hung Nguyen, Morris Chang

This study examines the impact of class-imbalanced data on deep learning models and proposes a technique for data balancing by generating synthetic data for the minority class. Unlike random-based oversampling, our method prioritizes balancing the informative regions by identifying high entropy samples. Generating well-placed synthetic data can enhance machine learning algorithms accuracy and efficiency, whereas poorly-placed ones may lead to higher misclassification rates. We introduce an algorithm that maximizes the probability of generating a synthetic sample in the correct region of its class by optimizing the class posterior ratio. Additionally, to maintain data topology, synthetic data are generated within each minority sample's neighborhood. Our experimental results on forty-one datasets demonstrate the superior performance of our technique in enhancing deep-learning models.

Hung Nguyen, An Le, Runfa Li et al.

3D Gaussian Splatting has emerged as a powerful approach in novel view synthesis, delivering rapid training and rendering but at the cost of an ever-growing set of Gaussian primitives that strains memory and bandwidth. We introduce AutoOpti3DGS, a training-time framework that automatically restrains Gaussian proliferation without sacrificing visual fidelity. The key idea is to feed the input images to a sequence of learnable Forward and Inverse Discrete Wavelet Transforms, where low-pass filters are kept fixed, high-pass filters are learnable and initialized to zero, and an auxiliary orthogonality loss gradually activates fine frequencies. This wavelet-driven, coarse-to-fine process delays the formation of redundant fine Gaussians, allowing 3DGS to capture global structure first and refine detail only when necessary. Through extensive experiments, AutoOpti3DGS requires just a single filter learning-rate hyper-parameter, integrates seamlessly with existing efficient 3DGS frameworks, and consistently produces sparser scene representations more compatible with memory or storage-constrained hardware.

Hung Nguyen, Runfa Li, An Le et al.

Sparse-view 3D Gaussian Splatting (3DGS) presents significant challenges in reconstructing high-quality novel views, as it often overfits to the widely-varying high-frequency (HF) details of the sparse training views. While frequency regularization can be a promising approach, its typical reliance on Fourier transforms causes difficult parameter tuning and biases towards detrimental HF learning. We propose DWTGS, a framework that rethinks frequency regularization by leveraging wavelet-space losses that provide additional spatial supervision. Specifically, we supervise only the low-frequency (LF) LL subbands at multiple DWT levels, while enforcing sparsity on the HF HH subband in a self-supervised manner. Experiments across benchmarks show that DWTGS consistently outperforms Fourier-based counterparts, as this LF-centric strategy improves generalization and reduces HF hallucinations.

An Le, Nehal Mehta, William Freeman et al.

In this study, we developed deep learning-based method to classify the type of surgery performed for epiretinal membrane (ERM) removal, either internal limiting membrane (ILM) removal or ERM-alone removal. Our model, based on the ResNet18 convolutional neural network (CNN) architecture, utilizes postoperative optical coherence tomography (OCT) center scans as inputs. We evaluated the model using both original scans and scans preprocessed with energy crop and wavelet denoising, achieving 72% accuracy on preprocessed inputs, outperforming the 66% accuracy achieved on original scans. To further improve accuracy, we integrated tunable wavelet units with two key adaptations: Orthogonal Lattice-based Wavelet Units (OrthLatt-UwU) and Perfect Reconstruction Relaxation-based Wavelet Units (PR-Relax-UwU). These units allowed the model to automatically adjust filter coefficients during training and were incorporated into downsampling, stride-two convolution, and pooling layers, enhancing its ability to distinguish between ERM-ILM removal and ERM-alone removal, with OrthLattUwU boosting accuracy to 76% and PR-Relax-UwU increasing performance to 78%. Performance comparisons showed that our AI model outperformed a trained human grader, who achieved only 50% accuracy in classifying the removal surgery types from postoperative OCT scans. These findings highlight the potential of CNN based models to improve clinical decision-making by providing more accurate and reliable classifications. To the best of our knowledge, this is the first work to employ tunable wavelets for classifying different types of ERM removal surgery.

An Le, Hung Nguyen, Sungbal Seo et al.

This work introduces a novel biorthogonal tunable wavelet unit constructed using a lifting scheme that relaxes both the orthogonality and equal filter length constraints, providing greater flexibility in filter design. The proposed unit enhances convolution, pooling, and downsampling operations, leading to improved image classification and anomaly detection in convolutional neural networks (CNN). When integrated into an 18-layer residual neural network (ResNet-18), the approach improved classification accuracy on CIFAR-10 by 2.12% and on the Describable Textures Dataset (DTD) by 9.73%, demonstrating its effectiveness in capturing fine-grained details. Similar improvements were observed in ResNet-34. For anomaly detection in the hazelnut category of the MVTec Anomaly Detection dataset, the proposed method achieved competitive and wellbalanced performance in both segmentation and detection tasks, outperforming existing approaches in terms of accuracy and robustness.

Hung Nguyen, An Le, Truong Nguyen

3D Gaussian Splatting (3DGS) has emerged as a powerful approach for novel view synthesis. However, the number of Gaussian primitives often grows substantially during training as finer scene details are reconstructed, leading to increased memory and storage costs. Recent coarse-to-fine strategies regulate Gaussian growth by modulating the frequency content of the ground-truth images. In particular, AutoOpti3DGS employs the learnable Discrete Wavelet Transform (DWT) to enable data-adaptive frequency modulation. Nevertheless, its modulation depth is limited by the 1-level DWT, and jointly optimizing wavelet regularization with 3D reconstruction introduces gradient competition that promotes excessive Gaussian densification. In this paper, we propose a multi-level DWT-based frequency modulation framework for 3DGS. By recursively decomposing the low-frequency subband, we construct a deeper curriculum that provides progressively coarser supervision during early training, consistently reducing Gaussian counts. Furthermore, we show that the modulation can be performed using only a single scaling parameter, rather than learning the full 2-tap high-pass filter. Experimental results on standard benchmarks demonstrate that our method further reduces Gaussian counts while maintaining competitive rendering quality.

Hung Nguyen, Runfa Li, An Le et al.

3D Gaussian Splatting (3DGS) has become a powerful representation for image-based object reconstruction, yet its performance drops sharply in sparse-view settings. Prior works address this limitation by employing diffusion models to repair corrupted renders, subsequently using them as pseudo ground truths for later optimization. While effective, such approaches incur heavy computation from the diffusion fine-tuning and repair steps. We present WaveletGaussian, a framework for more efficient sparse-view 3D Gaussian object reconstruction. Our key idea is to shift diffusion into the wavelet domain: diffusion is applied only to the low-resolution LL subband, while high-frequency subbands are refined with a lightweight network. We further propose an efficient online random masking strategy to curate training pairs for diffusion fine-tuning, replacing the commonly used, but inefficient, leave-one-out strategy. Experiments across two benchmark datasets, Mip-NeRF 360 and OmniObject3D, show WaveletGaussian achieves competitive rendering quality while substantially reducing training time.

An D. Le, Hung Nguyen, Melanie Tran et al.

This paper presents the first study to apply tunable wavelet units (UwUs) for 3D retinal layer segmentation from Optical Coherence Tomography (OCT) volumes. To overcome the limitations of conventional max-pooling, we integrate three wavelet-based downsampling modules, OrthLattUwU, BiorthLattUwU, and LS-BiorthLattUwU, into a motion-corrected MGU-Net architecture. These modules use learnable lattice filter banks to preserve both low- and high-frequency features, enhancing spatial detail and structural consistency. Evaluated on the Jacobs Retina Center (JRC) OCT dataset, our framework shows significant improvement in accuracy and Dice score, particularly with LS-BiorthLattUwU, highlighting the benefits of tunable wavelet filters in volumetric medical image segmentation.

An D. Le, Hung Nguyen, Sungbal Seo et al.

This work introduces a stop-band energy constraint for filters in orthogonal tunable wavelet units with a lattice structure, aimed at improving image classification and anomaly detection in CNNs, especially on texture-rich datasets. Integrated into ResNet-18, the method enhances convolution, pooling, and downsampling operations, yielding accuracy gains of 2.48% on CIFAR-10 and 13.56% on the Describable Textures dataset. Similar improvements are observed in ResNet-34. On the MVTec hazelnut anomaly detection task, the proposed method achieves competitive results in both segmentation and detection, outperforming existing approaches.

Huy Q. Ngo, Mingyu Guo, Hung Nguyen

Security vulnerabilities in Windows Active Directory (AD) systems are typically modeled using an attack graph and hardening AD systems involves an iterative workflow: security teams propose an edge to remove, and IT operations teams manually review these fixes before implementing the removal. As verification requires significant manual effort, we formulate an Adaptive Path Removal Problem to minimize the number of steps in this iterative removal process. In our model, a wizard proposes an attack path in each step and presents it as a set of multiple-choice options to the IT admin. The IT admin then selects one edge from the proposed set to remove. This process continues until the target $t$ is disconnected from source $s$ or the number of proposed paths reaches $B$. The model aims to optimize the human effort by minimizing the expected number of interactions between the IT admin and the security wizard. We first prove that the problem is $\mathcal{\#P}$-hard. We then propose a set of solutions including an exact algorithm, an approximate algorithm, and several scalable heuristics. Our best heuristic, called DPR, can operate effectively on larger-scale graphs compared to the exact algorithm and consistently outperforms the approximate algorithm across all graphs. We verify the effectiveness of our algorithms on several synthetic AD graphs and an AD attack graph collected from a real organization.

Hung Nguyen, Quang Qui-Vinh Nguyen, Khoi Nguyen et al.

Given an input video of a person and a new garment, the objective of this paper is to synthesize a new video where the person is wearing the specified garment while maintaining spatiotemporal consistency. Although significant advances have been made in image-based virtual try-on, extending these successes to video often leads to frame-to-frame inconsistencies. Some approaches have attempted to address this by increasing the overlap of frames across multiple video chunks, but this comes at a steep computational cost due to the repeated processing of the same frames, especially for long video sequences. To tackle these challenges, we reconceptualize video virtual try-on as a conditional video inpainting task, with garments serving as input conditions. Specifically, our approach enhances image diffusion models by incorporating temporal attention layers to improve temporal coherence. To reduce computational overhead, we propose ShiftCaching, a novel technique that maintains temporal consistency while minimizing redundant computations. Furthermore, we introduce the TikTokDress dataset, a new video try-on dataset featuring more complex backgrounds, challenging movements, and higher resolution compared to existing public datasets. Extensive experiments demonstrate that our approach outperforms current baselines, particularly in terms of video consistency and inference speed. The project page is available at https://swift-try.github.io/.

Mina Aghaei Dinani, Adrian Holzer, Hung Nguyen et al.

Fully distributed learning schemes such as Gossip Learning (GL) are gaining momentum due to their scalability and effectiveness even in dynamic settings. However, they often imply a high utilization of communication and computing resources, whose energy footprint may jeopardize the learning process, particularly on battery-operated IoT devices. To address this issue, we present Optimized Gossip Learning (OGL)}, a distributed training approach based on the combination of GL with adaptive optimization of the learning process, which allows for achieving a target accuracy while minimizing the energy consumption of the learning process. We propose a data-driven approach to OGL management that relies on optimizing in real-time for each node the number of training epochs and the choice of which model to exchange with neighbors based on patterns of node contacts, models' quality, and available resources at each node. Our approach employs a DNN model for dynamic tuning of the aforementioned parameters, trained by an infrastructure-based orchestrator function. We performed our assessments on two different datasets, leveraging time-varying random graphs and a measurement-based dynamic urban scenario. Results suggest that our approach is highly efficient and effective in a broad spectrum of network scenarios.

Gemini Team, Rohan Anil, Sebastian Borgeaud et al.

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.

Mingyu Guo, Jialiang Li, Aneta Neumann et al.

Active Directory is the default security management system for Windows domain networks. We study the shortest path edge interdiction problem for defending Active Directory style attack graphs. The problem is formulated as a Stackelberg game between one defender and one attacker. The attack graph contains one destination node and multiple entry nodes. The attacker's entry node is chosen by nature. The defender chooses to block a set of edges limited by his budget. The attacker then picks the shortest unblocked attack path. The defender aims to maximize the expected shortest path length for the attacker, where the expectation is taken over entry nodes. We observe that practical Active Directory attack graphs have small maximum attack path lengths and are structurally close to trees. We first show that even if the maximum attack path length is a constant, the problem is still $W[1]$-hard with respect to the defender's budget. Having a small maximum attack path length and a small budget is not enough to design fixed-parameter algorithms. If we further assume that the number of entry nodes is small, then we derive a fixed-parameter tractable algorithm. We then propose two other fixed-parameter algorithms by exploiting the tree-like features. One is based on tree decomposition and requires a small tree width. The other assumes a small number of splitting nodes (nodes with multiple out-going edges). Finally, the last algorithm is converted into a graph convolutional neural network based heuristic, which scales to larger graphs with more splitting nodes.

Hung Nguyen, Minh Nguyen, Long Pham et al.

In this paper, we develop a recommender system for a game that suggests potential items to players based on their interactive behaviors to maximize revenue for the game provider. Our approach is built on a reinforcement learning-based technique and is trained on an offline data set that is publicly available on an IEEE Big Data Cup challenge. The limitation of the offline data set and the curse of high dimensionality pose significant obstacles to solving this problem. Our proposed method focuses on improving the total rewards and performance by tackling these main difficulties. More specifically, we utilized sparse PCA to extract important features of user behaviors. Our Q-learning-based system is then trained from the processed offline data set. To exploit all possible information from the provided data set, we cluster user features to different groups and build an independent Q-table for each group. Furthermore, to tackle the challenge of unknown formula for evaluation metrics, we design a metric to self-evaluate our system's performance based on the potential value the game provider might achieve and a small collection of actual evaluation metrics that we obtain from the live scoring environment. Our experiments show that our proposed metric is consistent with the results published by the challenge organizers. We have implemented the proposed training pipeline, and the results show that our method outperforms current state-of-the-art methods in terms of both total rewards and training speed. By addressing the main challenges and leveraging the state-of-the-art techniques, we have achieved the best public leaderboard result in the challenge. Furthermore, our proposed method achieved an estimated score of approximately 20% better and can be trained faster by 30 times than the best of the current state-of-the-art methods.

Hung Nguyen, Morris Chang

To achieve high performance of a machine learning (ML) task, a deep learning-based model must implicitly capture the entire distribution from data. Thus, it requires a huge amount of training samples, and data are expected to fully present the real distribution, especially for high dimensional data, e.g., images, videos. In practice, however, data are usually collected with a diversity of styles, and several of them have insufficient number of representatives. This might lead to uncertainty in models' prediction, and significantly reduce ML task performance. In this paper, we provide a comprehensive study on this problem by looking at model uncertainty. From this, we derive a simple but efficient technique to improve performance of state-of-the-art deep learning models. Specifically, we train auxiliary models which are able to complement state-of-the-art model uncertainty. As a result, by assembling these models, we can significantly improve the ML task performance for types of data mentioned earlier. While slightly improving ML classification accuracy on benchmark datasets (e.g., 0.2% on MNIST), our proposed method significantly improves on limited data (i.e., 1.3% on Eardrum and 3.5% on ChestXray).

Hung Nguyen, Chanho Kim, Fuxin Li

Transformers have recently been popular for learning and inference in the spatial-temporal domain. However, their performance relies on storing and applying attention to the feature tensor of each frame in video. Hence, their space and time complexity increase linearly as the length of video grows, which could be very costly for long videos. We propose a novel visual memory network architecture for the learning and inference problem in the spatial-temporal domain. We maintain a fixed set of memory slots in our memory network and propose an algorithm based on Gumbel-Softmax to learn an adaptive strategy to update this memory. Finally, this architecture is benchmarked on the video object segmentation (VOS) and video prediction problems. We demonstrate that our memory architecture achieves state-of-the-art results, outperforming transformer-based methods on VOS and other recent methods on video prediction while maintaining constant memory capacity independent of the sequence length.

Hung Nguyen, Xuejian Wang, Leman Akoglu

Given a labeled dataset that contains a rare (or minority) class of of-interest instances, as well as a large class of instances that are not of interest, how can we learn to recognize future of-interest instances over a continuous stream? We introduce RaRecognize, which (i) estimates a general decision boundary between the rare and the majority class, (ii) learns to recognize individual rare subclasses that exist within the training data, as well as (iii) flags instances from previously unseen rare subclasses as newly emerging. The learner in (i) is general in the sense that by construction it is dissimilar to the specialized learners in (ii), thus distinguishes minority from the majority without overly tuning to what is seen in the training data. Thanks to this generality, RaRecognize ignores all future instances that it labels as majority and recognizes the recurrent as well as emerging rare subclasses only. This saves effort at test time as well as ensures that the model size grows moderately over time as it only maintains specialized minority learners. Through extensive experiments, we show that RaRecognize outperforms state-of-the art baselines on three real-world datasets that contain corporate-risk and disaster documents as rare classes.

Gaetano Manzo, Sebastian Otalora, Marco Ajmone Marsan et al.

Opportunistic communications are expected to playa crucial role in enabling context-aware vehicular services. A widely investigated opportunistic communication paradigm for storing a piece of content probabilistically in a geographica larea is Floating Content (FC). A key issue in the practical deployment of FC is how to tune content replication and caching in a way which achieves a target performance (in terms of the mean fraction of users possessing the content in a given region of space) while minimizing the use of bandwidth and host memory. Fully distributed, distance-based approaches prove highly inefficient, and may not meet the performance target,while centralized, model-based approaches do not perform well in realistic, inhomogeneous settings. In this work, we present a data-driven centralized approach to resource-efficient, QoS-aware dynamic management of FC.We propose a Deep Learning strategy, which employs a Convolutional Neural Network (CNN) to capture the relationships between patterns of users mobility, of content diffusion and replication, and FC performance in terms of resource utilization and of content availability within a given area. Numerical evaluations show the effectiveness of our approach in deriving strategies which efficiently modulate the FC operation in space and effectively adapt to mobility pattern changes over time.

Hung Nguyen, Di Zhuang, Pei-Yuan Wu et al.

Protecting sensitive information against data exploiting attacks is an emerging research area in data mining. Over the past, several different methods have been introduced to protect individual privacy from such attacks while maximizing data-utility of the application. However, these existing techniques are not sufficient to effectively protect data owner privacy, especially in the scenarios that utilize visualizable data (e.g. images, videos) or the applications that require heavy computations for implementation. To address these problems, we propose a new dimension reduction-based method for privacy preservation. Our method generates dimension-reduced data for performing machine learning tasks and prevents a strong adversary from reconstructing the original data. We first introduce a theoretical approach to evaluate dimension reduction-based privacy preserving mechanisms, then propose a non-linear dimension reduction framework motivated by state-of-the-art neural network structures for privacy preservation. We conducted experiments over three different face image datasets (AT&T, YaleB, and CelebA), and the results show that when the number of dimensions is reduced to seven, we can achieve the accuracies of 79%, 80%, and 73% respectively and the reconstructed images are not recognizable to naked human eyes.

Hung Nguyen, Vu Tran, Tung Nguyen et al.

In apprenticeship learning (AL), agents learn by watching or acquiring human demonstrations on some tasks of interest. However, the lack of human demonstrations in novel tasks where they may not be a human expert yet, or when it is too expensive and/or time consuming to acquire human demonstrations motivated a new algorithm: Apprenticeship bootstrapping (ABS). The basic idea is to learn from demonstrations on sub-tasks then autonomously bootstrap a model on the main, more complex, task. The original ABS used inverse reinforcement learning (ABS-IRL). However, the approach is not suitable for continuous action spaces. In this paper, we propose ABS via Deep learning (ABS-DL). It is first validated in a simulation environment on an aerial and ground coordination scenario, where an Unmanned Aerial Vehicle (UAV) is required to maintain three Unmanned Ground Vehicles (UGVs) within a field of view of the UAV 's camera (FoV). Moving a machine learning algorithm from a simulation environment to an actual physical platform is challenging because `mistakes' made by the algorithm while learning could lead to the damage of the platform. We then take this extra step to test the algorithm in a physical environment. We propose a safety-net as a protection layer to ensure that the autonomy of the algorithm in learning does not compromise the safety of the platform. The tests of ABS-DL in the real environment can guarantee a damage-free, collision avoidance behaviour of autonomous bodies. The results show that performance of the proposed approach is comparable to that of a human, and competitive to the traditional approach using expert demonstrations performed on the composite task. The proposed safety-net approach demonstrates its advantages when it enables the UAV to operate more safely under the control of the ABS-DL algorithm.

Aya Hussein, Leo Ghignone, Tung Nguyen et al.

Swarm systems consist of large numbers of robots that collaborate autonomously. With an appropriate level of human control, swarm systems could be applied in a variety of contexts ranging from search-and-rescue situations to Cyber defence. The two decision making cycles of swarms and humans operate on two different time-scales, where the former is normally orders of magnitude faster than the latter. Closing the loop at the intersection of these two cycles will create fast and adaptive human-swarm teaming networks. This paper brings desperate pieces of the ground work in this research area together to review this multidisciplinary literature. We conclude with a framework to synthesize the findings and summarize the multi-modal indicators needed for closed-loop human-swarm adaptive systems.