A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications

Journal Of Big Data, Journal Year: 2023, Volume and Issue: 10(1)

Published: April 14, 2023

Abstract Data scarcity is a major challenge when training deep learning (DL) models. DL demands large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate train frameworks. Usually, manual labeling needed provide labeled data, which typically involves human annotators with vast background knowledge. This annotation process costly, time-consuming, and error-prone. every framework fed by significant automatically learn representations. Ultimately, larger would generate better model its performance also application dependent. issue the main barrier for dismissing use DL. Having sufficient first step toward any successful trustworthy application. paper presents holistic survey on state-of-the-art techniques deal models overcome three challenges including small, imbalanced datasets, lack generalization. starts listing techniques. Next, types architectures are introduced. After that, solutions address listed, such as Transfer Learning (TL), Self-Supervised (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these were followed some related tips about acquisition prior purposes, well recommendations ensuring trustworthiness dataset. The ends list that suffer from scarcity, several alternatives proposed in order more each Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, Cybersecurity. To best authors’ knowledge, this review offers comprehensive overview strategies tackle

Language: Английский

---

Physics-Guided, Physics-Informed, and Physics-Encoded Neural Networks and Operators in Scientific Computing: Fluid and Solid Mechanics

Journal of Computing and Information Science in Engineering, Journal Year: 2024, Volume and Issue: 24(4)

Published: Jan. 8, 2024

Abstract Advancements in computing power have recently made it possible to utilize machine learning and deep push scientific forward a range of disciplines, such as fluid mechanics, solid materials science, etc. The incorporation neural networks is particularly crucial this hybridization process. Due their intrinsic architecture, conventional cannot be successfully trained scoped when data are sparse, which the case many engineering domains. Nonetheless, provide foundation respect physics-driven or knowledge-based constraints during training. Generally speaking, there three distinct network frameworks enforce underlying physics: (i) physics-guided (PgNNs), (ii) physics-informed (PiNNs), (iii) physics-encoded (PeNNs). These methods advantages for accelerating numerical modeling complex multiscale multiphysics phenomena. In addition, recent developments operators (NOs) add another dimension these new simulation paradigms, especially real-time prediction systems required. All models also come with own unique drawbacks limitations that call further fundamental research. This study aims present review four (i.e., PgNNs, PiNNs, PeNNs, NOs) used state-of-the-art architectures applications reviewed, discussed, future research opportunities presented terms improving algorithms, considering causalities, expanding applications, coupling solvers.

Language: Английский

---

Deep learning in computational mechanics: a review

Computational Mechanics, Journal Year: 2024, Volume and Issue: 74(2), P. 281 - 331

Published: Jan. 13, 2024

Abstract The rapid growth of deep learning research, including within the field computational mechanics, has resulted in an extensive and diverse body literature. To help researchers identify key concepts promising methodologies this field, we provide overview deterministic mechanics. Five main categories are identified explored: simulation substitution, enhancement, discretizations as neural networks, generative approaches, reinforcement learning. This review focuses on methods rather than applications for thereby enabling to explore more effectively. As such, is not necessarily aimed at with knowledge learning—instead, primary audience verge entering or those attempting gain discussed are, therefore, explained simple possible.

Language: Английский

---

A complete Physics-Informed Neural Network-based framework for structural topology optimization

Computer Methods in Applied Mechanics and Engineering, Journal Year: 2023, Volume and Issue: 417, P. 116401 - 116401

Published: Sept. 9, 2023

Physics-Informed Neural Networks (PINNs) have recently gained increasing attention in the field of topology optimization. The fusion deep learning and optimization has emerged as a prominent area insightful research, where minimization loss function neural networks can be comparable to objective Inspired by concepts PINNs, this paper proposes novel framework, 'Complete Network-based Topology Optimization (CPINNTO)', address various challenges optimization, particularly related structural key innovation proposed framework lies introducing first complete machine-learning-based through integration two distinct PINNs. Herein, Deep Energy Method (DEM) PINN is implemented determine deformation state corresponding structures numerically. In addition, derivation with respect design variables replaced automatic differentiation sensitivity-analysis (S-PINN). feasibility potential CPINNTO been assessed several case studies while highlighting strengths limitations utilizing PINNs Subsequent findings indicate that achieve optimal topologies without labeled data nor FEA. numerical examples demonstrate capable stably obtaining for applications, including compliance problems, multi-constrained three-dimensional problems. Resulting designs exhibit favorable values obtained via density-based summary, opens up interesting possibilities

Language: Английский

---

Exploring energy minimization to model strain localization as a strong discontinuity using Physics Informed Neural Networks

Computer Methods in Applied Mechanics and Engineering, Journal Year: 2025, Volume and Issue: 436, P. 117724 - 117724

Published: Jan. 11, 2025

Language: Английский

---

Energy-based physics-informed neural network for frictionless contact problems under large deformation

Computer Methods in Applied Mechanics and Engineering, Journal Year: 2025, Volume and Issue: 437, P. 117787 - 117787

Published: Jan. 30, 2025

Language: Английский

---

A comprehensive review of fiber-reinforced polymer-matrix composites under low-velocity impact

Mechanics of Advanced Materials and Structures, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 39

Published: Feb. 6, 2025

Language: Английский

---

Energy-based PINNs using the element integral approach and their enhancement for solid mechanics problems

International Journal of Solids and Structures, Journal Year: 2025, Volume and Issue: unknown, P. 113315 - 113315

Published: Feb. 1, 2025

Language: Английский

---

PINNs-MPF: A Physics-Informed Neural Network framework for Multi-Phase-Field simulation of interface dynamics

Engineering Analysis with Boundary Elements, Journal Year: 2025, Volume and Issue: 176, P. 106200 - 106200

Published: April 3, 2025

Language: Английский

---

A novel physics-informed neural networks approach (PINN-MT) to solve mass transfer in plant cells during drying

Biosystems Engineering, Journal Year: 2023, Volume and Issue: 230, P. 219 - 241

Published: May 11, 2023

Language: Английский

---