Searcharxiv⌕ Search

arXiv subjects

Ton Viet Ta

Publications and source records attributed to Ton Viet Ta.

At least 19 recordsLinked to original sources
arXiv2026

BioSEN: A Bio-acoustic Signal Enhancement Network for Animal Vocalizations

Most work in audio enhancement targets human speech, while bioacoustics is less studied due to noisy recordings and the distinct traits of animal sounds. To fill this gap, we adapt speech enhancement methods and build BioSEN, a model made for bioacoustic signals. BioSEN has three modules: a multi-scale dual-axis attention unit for time-frequency feature extraction, a bio-harmonic multi-scale enhancement unit for capturing harmonic structures, and an energy-adaptive gating connection unit that uses frequency weights to keep vocalizations from being removed as noise. Tests on three bioacoustic datasets show that BioSEN matches or exceeds state-of-the-art speech enhancement models while using far less computation. These results show BioSEN's strength for bioacoustic audio enhancement and its promise for biodiversity monitoring and conservation.

cs.SD↗
arXiv2026

Emergent spatial organization of competing species under environmental stress and cooperation

Understanding how species persist under interacting stressors is a central challenge in ecology. We develop a spatially explicit reaction-diffusion framework to investigate competing species in landscapes shaped by climate variability, pollution, resource heterogeneity, and cooperation. Here, temperature follows low-frequency oscillations, while pollution and resources diffuse from localized sources. Growth is governed by a dynamic carrying capacity integrating abiotic stress with an endogenous, pollution-sensitive cooperation field. Numerical simulations reveal the spontaneous emergence of persistent spatial organization, including dominance segregation and stable competitive boundaries. Quantitative analyses-using boundary geometry, fractal dimension, and spatial entropy-demonstrate a transition from intermixed initial states to low-complexity, quasi-stationary configurations. Coexistence occurs through distinct strategies: one species occupies more area, while the other maintains higher local densities. Cooperation enhances resilience but collapses in polluted zones, creating heterogeneous "social buffering." We further introduce a hybrid inverse modeling framework using a Swin Transformer to infer high-dimensional parameters from only two temporal snapshots. Trained on synthetic data, the model accurately recovers demographic, diffusive, and environmental-sensitivity parameters. While it achieves reliable short-term spatial predictions, long-term forecasts diverge due to the intrinsic sensitivity of nonlinear systems. This unified framework links sparse observations to mechanistic dynamics, advancing biodiversity forecasting under accelerating global change.

q-bio.PE↗
arXiv2026

Bifurcation Analysis of a Predator-Prey Model with Allee Effect and Cooperative Hunting

We propose a novel predator-prey model that integrate two ecologically significant mechanisms: the Allee effect in the prey population and cooperative hunting behavior among predators. Building upon the Rosenzweig-MacArthur framework, our model modifies the prey growth term to incorporate the Allee effect and introduces a nonlinear functional response reflecting predator cooperation. We establish the existence and boundedness of global solutions for the system and analyze the local and global stability of its equilibria. In addition, we perform a comprehensive bifurcation analysis, including transcritical, saddle-node, Hopf, and heteroclinic bifurcations, to explore how system dynamics change with key parameters. These results reveal rich and biologically relevant behaviors, such as multiple equilibria, transitions in stability, and the emergence of complex dynamical patterns.

math.DS↗
arXiv2025

Modeling Predator-Prey Dynamics with Stochastic Differential Equations: Patterns of Collective Hunting and Nonlinear Predation Effects

We investigate predator-prey school interactions in aquatic environments using a stochastic differential equation (SDE)-based, particle-level model that incorporates attraction, repulsion, alignment, and environmental noise. Two predation strategies-center attack and nearest attack-are examined to assess their effects on prey survival, predator efficiency, and group dynamics. Simulations reveal diverse emergent behaviors such as prey dispersal and regrouping, oscillatory predation with collective defense, and predator encirclement. Results show that collective hunting enhances capture efficiency compared to solitary attacks, but benefits diminish beyond a critical predator group size due to intra-predator competition. This work provides new insights into cooperative predation and introduces a generalizable SDE framework for analyzing predator-prey interactions.

q-bio.PE↗
arXiv2025

Towards High-Fidelity and Controllable Bioacoustic Generation via Enhanced Diffusion Learning

Generative modeling offers new opportunities for bioacoustics, enabling the synthesis of realistic animal vocalizations that could support biomonitoring efforts and supplement scarce data for endangered species. However, directly generating bird call waveforms from noisy field recordings remains a major challenge. We propose BirdDiff, a generative framework designed to synthesize bird calls from a noisy dataset of 12 wild bird species. The model incorporates a "zeroth layer" stage for multi-scale adaptive bird-call enhancement, followed by a diffusion-based generator conditioned on three modalities: Mel-frequency cepstral coefficients, species labels, and textual descriptions. The enhancement stage improves signal-to-noise ratio (SNR) while minimizing spectral distortion, achieving the highest SNR gain (+10.45 dB) and lowest Itakura-Saito Distance (0.54) compared to three widely used non-training enhancement methods. We evaluate BirdDiff against a baseline generative model, DiffWave. Our method yields substantial improvements in generative quality metrics: Fréchet Audio Distance (0.590 to 0.213), Jensen-Shannon Divergence (0.259 to 0.226), and Number of Statistically-Different Bins (7.33 to 5.58). To assess species-specific detail preservation, we use a ResNet50 classifier trained on the original dataset to identify generated samples. Classification accuracy improves from 35.9% (DiffWave) to 70.1% (BirdDiff), with 8 of 12 species exceeding 70% accuracy. These results demonstrate that BirdDiff enables high-fidelity, controllable bird call generation directly from noisy field recordings.

cs.SD↗
arXiv2025

Deep Learning for Automated Identification of Vietnamese Timber Species: A Tool for Ecological Monitoring and Conservation

Accurate identification of wood species plays a critical role in ecological monitoring, biodiversity conservation, and sustainable forest management. Traditional classification approaches relying on macroscopic and microscopic inspection are labor-intensive and require expert knowledge. In this study, we explore the application of deep learning to automate the classification of ten wood species commonly found in Vietnam. A custom image dataset was constructed from field-collected wood samples, and five state-of-the-art convolutional neural network architectures--ResNet50, EfficientNet, MobileViT, MobileNetV3, and ShuffleNetV2--were evaluated. Among these, ShuffleNetV2 achieved the best balance between classification performance and computational efficiency, with an average accuracy of 99.29\% and F1-score of 99.35\% over 20 independent runs. These results demonstrate the potential of lightweight deep learning models for real-time, high-accuracy species identification in resource-constrained environments. Our work contributes to the growing field of ecological informatics by providing scalable, image-based solutions for automated wood classification and forest biodiversity assessment.

cs.CV↗
arXiv2025

Mathematical Models for Fish Schooling

This note reviews our mathematical models for fish schooling, considered in free space, and in space with obstacle and food resource. These models are performed by stochastic differential equations or stochastic partial differential equations. We then present an example for the model in the last case.

cond-mat.stat-mech↗
arXiv2025

Quantifying Fish School Fragmentation under Predation Using Stochastic Differential Equations

This study builds upon our previously proposed stochastic differential equation (SDE)-based model to further investigate fish school fragmentation under predation. Specifically, we explore structural dynamics by incorporating graph-theoretic metrics--namely, the number of connected components--to quantify changes in prey school organization. Two quantitative indicators, first split time and final component count, are introduced to assess the timing and extent of group disintegration. Sensitivity analyses are performed on key parameters to evaluate their influence on group stability under nearest attack and center attack strategies. We independently examine the effect of environmental noise on fish school cohesion. Simulation results show that parameter changes impact fish school fragmentation differently under the two predation strategies. High environmental noise also makes it difficult for the school to stay cohesive. This framework provides a structured and quantitative basis for assessing how fish schools respond to different predation strategies and environmental noise levels.

q-bio.QM↗
arXiv2025

Stochastic forest transition model dynamics and parameter estimation via deep learning

Forest transitions, characterized by dynamic shifts between forest, agricultural, and abandoned lands, are complex phenomena. This study developed a stochastic differential equation model to capture the intricate dynamics of these transitions. We established the existence of global positive solutions for the model and conducted numerical analyses to assess the impact of model parameters on deforestation incentives. To address the challenge of parameter estimation, we proposed a novel deep learning approach that estimates all model parameters from a single sample containing time-series observations of forest and agricultural land proportions. This innovative approach enables us to understand forest transition dynamics and deforestation trends at any future time.

stat.ML↗
arXiv2024

Deep learning-based method for weather forecasting: A case study in Itoshima

Accurate weather forecasting is of paramount importance for a wide range of practical applications, drawing substantial scientific and societal interest. However, the intricacies of weather systems pose substantial challenges to accurate predictions. This research introduces a multilayer perceptron model tailored for weather forecasting in Itoshima, Kyushu, Japan. Our meticulously designed architecture demonstrates superior performance compared to existing models, surpassing benchmarks such as Long Short-Term Memory and Recurrent Neural Networks.

cs.LG↗
arXiv2023

MPSA-DenseNet: A novel deep learning model for English accent classification

This paper presents three innovative deep learning models for English accent classification: Multi-DenseNet, PSA-DenseNet, and MPSE-DenseNet, that combine multi-task learning and the PSA module attention mechanism with DenseNet. We applied these models to data collected from six dialects of English across native English speaking regions (Britain, the United States, Scotland) and nonnative English speaking regions (China, Germany, India). Our experimental results show a significant improvement in classification accuracy, particularly with MPSA-DenseNet, which outperforms all other models, including DenseNet and EPSA models previously used for accent identification. Our findings indicate that MPSA-DenseNet is a highly promising model for accurately identifying English accents.

cs.CL↗
arXiv2023

Dynamics of infectious diseases in predator-prey populations: a stochastic model, sustainability, and invariant measure

This paper introduces an innovative model for infectious diseases in predator-prey populations. We not only prove the existence of global non-negative solutions but also establish essential criteria for the system's decline and sustainability. Furthermore, we demonstrate the presence of a Borel invariant measure, adding a new dimension to our understanding of the system. To illustrate the practical implications of our findings, we present numerical results. With our model's comprehensive approach, we aim to provide valuable insights into the dynamics of infectious diseases and their impact on predator-prey populations.

q-bio.PE↗
arXiv2023

A Geometrical Structure for Predator-Avoidance Fish Schooling

This paper conducts a numerical study of a geometrical structure called $ε$-school for predator-avoidance fish schools, based on our previous mathematical model. Our results show that during a predator attack, the number of $ε$-school increases from one to a certain value. After the attack, the number of $ε$-school decreases in the first two predator-avoidance patterns, but continues to increase in the third pattern. A constant value for the number of the $ε$-school is observed in the last pattern. These suggests that when the predator is approaching, each individual in the school focuses more on avoiding the predator, rather than on interacting with its schoolmates. Such a trait is in agreement with real-life behavior in the natural ecosystem.

q-bio.PE↗
arXiv2022

A Stochastic Differential Equation Model for Predator-Avoidance Fish Schooling

This paper presents a system of stochastic differential equations (SDEs) as mathematical model to describe the spatial-temporal dynamics of predator-prey system in an artificial aquatic environment with schooling behavior imposed upon the associated prey. The proposed model follows the particle-like approach where interactions among the associated units are manifested through combination of attractive and repulsive forces analogous to the ones occurred in molecular physics. Two hunting tactics of the predator are proposed and integrated into the general model, namely the center-attacking and the nearest-attacking strategy. Emphasis is placed upon demonstrating the capacity of the proposed model in: (i) discovering the predator-avoidance patterns of the schooling prey, and (ii) showing the benefit of constituting large prey school in better escaping the predator's attack. Based on numerical simulations upon the proposed model, four predator-avoidance patterns of the schooling prey are discovered, namely Split and Reunion, Split and Separate into Two Groups, Scattered, and Maintain Formation and Distance. The proposed model also successfully demonstrates the benefit of constituting large group of schooling prey in mitigating predation risk. Such findings are in agreement with real-life observations of the natural aquatic ecosystem, hence confirming the validity and exactitude of the proposed model.

math.DS↗
arXiv2017

Stability of a one-predator two-prey system governed by nonautonomous differential equations

A non-periodic version of the one-predator two-prey system model presented in [L.T.H. Nguyen, Q.H. Ta, T.V. T\d{a}, Existence and stability of periodic solutions of a Lotka-Volterra system, SICE International Symposium on Control Systems, Tokyo, Japan, 712-4 (2015) 1-6] is considered. First, we prove existence of unique positive solutions to the model. Second, we show existence of an invariant set, which suggests the survival of all species in the system. On the other hand, we show that when the densities of two prey species are quite small, the predator falls into decay. Third, we explore global asymptotic stability of the system by using the Lyapunov function method. Finally, some numerical examples are given to illustrate our results.

math.DS↗
arXiv2017

A stochastic differential equation model for foraging behavior of fish schools

We present a novel model of stochastic differential equations for foraging behavior of fish schools in space including obstacles. We then study the model numerically. Three configurations of space with different locations of food resource are considered. In the first configuration, fish move in free but limited space. All individuals can find food almost surely. In the second and third configurations, fish move in limited space with one or two obstacles. Our results reveal that on one hand, when school size increases, so does the probability of foraging success. On the other hand, when it exceeds an optimal value, the probability decreases. In all configurations, fish always keep a school structure through the process of foraging.

math.PR↗
arXiv2017

Dynamical system for animal coat pattern model

We construct a dynamical system for a reaction diffusion system due to Murray, which relies on the use of the Thomas system nonlinearities and describes the formation of animal coat patterns. First, we prove existence and uniqueness of global positive strong solutions to the system by using semigroup methods. Second, we show that the solutions are continuously dependent on initial values. Third, we show that the dynamical system enjoys exponential attractors whose fractal dimensions can be estimated. Finally, we give a numerical example.

math.AP↗