Computational Engineering, Finance, and Science

New submissions

• New submissions
• Cross-lists
• Replacements

New submissions for Wed, 24 Apr 24

[1]  arXiv:2404.14591 [pdf, other]

Title: Predicting the Temporal Dynamics of Prosthetic Vision

Authors: Yuchen Hou, Laya Pullela, Jiaxin Su, Sriya Aluru, Shivani Sista, Xiankun Lu, Michael Beyeler

Subjects: Computational Engineering, Finance, and Science (cs.CE)

Retinal implants are a promising treatment option for degenerative retinal disease. While numerous models have been developed to simulate the appearance of elicited visual percepts ("phosphenes"), these models often either focus solely on spatial characteristics or inadequately capture the complex temporal dynamics observed in clinical trials, which vary heavily across subjects and stimulus conditions. Here we introduce two computational models designed to accurately predict phosphene fading and persistence under varying stimulus conditions, cross-validated on behavioral data reported by nine users of the Argus II Retinal Prosthesis System. Both models segment the time course of phosphene perception into discrete intervals, decomposing phosphene fading and persistence into either sinusoidal or exponential components. Our spectral model demonstrates state-of-the-art predictions of phosphene intensity over time (r = 0.7 across all participants). Overall, this study lays the groundwork for enhancing prosthetic vision by improving our understanding of phosphene temporal dynamics.

[2]  arXiv:2404.14984 [pdf, other]

Title: Surface profile recovery from electromagnetic field with physics--informed neural networks

Authors: Yuxuan Chen, Ce Wang, Yuan Hui, Mark Spivack

Subjects: Computational Engineering, Finance, and Science (cs.CE)

Physics--informed neural networks (PINN) have shown their potential in solving both direct and inverse problems of partial differential equations. In this paper, we introduce a PINN-based deep learning approach to reconstruct one-dimensional rough surfaces from field data illuminated by an electromagnetic incident wave. In the proposed algorithm, the rough surface is approximated by a neural network, with which the spatial derivatives of surface function can be obtained via automatic differentiation and then the scattered field can be calculated via the method of moments. The neural network is trained by minimizing the loss between the calculated and the observed field data. Furthermore, the proposed method is an unsupervised approach, independent of any surface data, rather only the field data is used. Both TE field (Dirichlet boundary condition) and TM field (Neumann boundary condition) are considered. Two types of field data are used here: full scattered field data and phaseless total field data. The performance of the method is verified by testing with Gaussian-correlated random rough surfaces. Numerical results demonstrate that the PINN-based method can recover rough surfaces with great accuracy and is robust with respect to a wide range of problem regimes.

[3]  arXiv:2404.15207 [pdf, other]

Title: Simulation-Free Determination of Microstructure Representative Volume Element Size via Fisher Scores

Authors: Wei Liu, Satyajit Mojumder, Wing Kam Liu, Wei Chen, Daniel W. Apley

Journal-ref: APL Mach. Learn. 2(2): 026101 (2024)

Subjects: Computational Engineering, Finance, and Science (cs.CE); Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG); Applications (stat.AP)

A representative volume element (RVE) is a reasonably small unit of microstructure that can be simulated to obtain the same effective properties as the entire microstructure sample. Finite element (FE) simulation of RVEs, as opposed to much larger samples, saves computational expense, especially in multiscale modeling. Therefore, it is desirable to have a framework that determines RVE size prior to FE simulations. Existing methods select the RVE size based on when the FE-simulated properties of samples of increasing size converge with insignificant statistical variations, with the drawback that many samples must be simulated. We propose a simulation-free alternative that determines RVE size based only on a micrograph. The approach utilizes a machine learning model trained to implicitly characterize the stochastic nature of the input micrograph. The underlying rationale is to view RVE size as the smallest moving window size for which the stochastic nature of the microstructure within the window is stationary as the window moves across a large micrograph. For this purpose, we adapt a recently developed Fisher score-based framework for microstructure nonstationarity monitoring. Because the resulting RVE size is based solely on the micrograph and does not involve any FE simulation of specific properties, it constitutes an RVE for any property of interest that solely depends on the microstructure characteristics. Through numerical experiments of simple and complex microstructures, we validate our approach and show that our selected RVE sizes are consistent with when the chosen FE-simulated properties converge.

Cross-lists for Wed, 24 Apr 24

[4]  arXiv:2404.14433 (cross-list from cs.LG) [pdf, other]

Title: KATO: Knowledge Alignment and Transfer for Transistor Sizing of Different Design and Technology

Authors: Wei W. Xing, Weijian Fan, Zhuohua Liu, Yuan Yao, Yuanqi Hu

Comments: 6 pages, received by DAC2024

Subjects: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE)

Automatic transistor sizing in circuit design continues to be a formidable challenge. Despite that Bayesian optimization (BO) has achieved significant success, it is circuit-specific, limiting the accumulation and transfer of design knowledge for broader applications. This paper proposes (1) efficient automatic kernel construction, (2) the first transfer learning across different circuits and technology nodes for BO, and (3) a selective transfer learning scheme to ensure only useful knowledge is utilized. These three novel components are integrated into BO with Multi-objective Acquisition Ensemble (MACE) to form Knowledge Alignment and Transfer Optimization (KATO) to deliver state-of-the-art performance: up to 2x simulation reduction and 1.2x design improvement over the baselines.

[5]  arXiv:2404.14688 (cross-list from cs.LG) [pdf, other]

Title: FMint: Bridging Human Designed and Data Pretrained Models for Differential Equation Foundation Model

Authors: Zezheng Song, Jiaxin Yuan, Haizhao Yang

Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Dynamical Systems (math.DS); Numerical Analysis (math.NA)

Human-designed algorithms have long been fundamental in solving a variety of scientific and engineering challenges. Recently, data-driven deep learning methods have also risen to prominence, offering innovative solutions across numerous scientific fields. While traditional algorithms excel in capturing the core aspects of specific problems, they often lack the flexibility needed for varying problem conditions due to the absence of specific data. Conversely, while data-driven approaches utilize vast datasets, they frequently fall short in domain-specific knowledge. To bridge these gaps, we introduce \textbf{FMint} (Foundation Model based on Initialization), a generative pre-trained model that synergizes the precision of human-designed algorithms with the adaptability of data-driven methods. This model is specifically engineered for high-accuracy simulation of dynamical systems. Starting from initial trajectories provided by conventional methods, FMint quickly delivers highly accurate solutions. It incorporates in-context learning and has been pre-trained on a diverse corpus of 500,000 dynamical systems, showcasing exceptional generalization across a broad spectrum of real-world applications. By effectively combining algorithmic rigor with data-driven flexibility, FMint sets the stage for the next generation of scientific foundation models, tackling complex problems with both efficiency and high accuracy.

[6]  arXiv:2404.14858 (cross-list from quant-ph) [pdf, other]

Title: A resource-efficient variational quantum algorithm for mRNA codon optimization

Authors: Hongfeng Zhang, Aritra Sarkar, Koen Bertels

Comments: 11 pages, 4 figures

Subjects: Quantum Physics (quant-ph); Computational Engineering, Finance, and Science (cs.CE)

Optimizing the mRNA codon has an essential impact on gene expression for a specific target protein. It is an NP-hard problem; thus, exact solutions to such optimization problems become computationally intractable for realistic problem sizes on both classical and quantum computers. However, approximate solutions via heuristics can substantially impact the application they enable. Quantum approximate optimization is an alternative computation paradigm promising for tackling such problems. Recently, there has been some research in quantum algorithms for bioinformatics, specifically for mRNA codon optimization. This research presents a denser way to encode codons for implementing mRNA codon optimization via the variational quantum eigensolver algorithms on a gate-based quantum computer. This reduces the qubit requirement by half compared to the existing quantum approach, thus allowing longer sequences to be executed on existing quantum processors. The performance of the proposed algorithm is evaluated by comparing its results to exact solutions, showing well-matching results.

[7]  arXiv:2404.14958 (cross-list from math.NA) [pdf, ps, other]

Title: Saving proof-of-work by hierarchical block structure

Authors: Valdemar Melicher

Subjects: Numerical Analysis (math.NA); Computational Engineering, Finance, and Science (cs.CE); Cryptography and Security (cs.CR); Computers and Society (cs.CY)

We argue that the current POW based consensus algorithm of the Bitcoin network suffers from a fundamental economic discrepancy between the real world transaction (txn) costs incurred by miners and the wealth that is being transacted. Put simply, whether one transacts 1 satoshi or 1 bitcoin, the same amount of electricity is needed when including this txn into a block. The notorious Bitcoin blockchain problems such as its high energy usage per txn or its scalability issues are, either partially or fully, mere consequences of this fundamental economic inconsistency. We propose making the computational cost of securing the txns proportional to the wealth being transferred, at least temporarily.
First, we present a simple incentive based model of Bitcoin's security. Then, guided by this model, we augment each txn by two parameters, one controlling the time spent securing this txn and the second determining the fraction of the network used to accomplish this. The current Bitcoin txns are naturally embedded into this parametrized space. Then we introduce a sequence of hierarchical block structures (HBSs) containing these parametrized txns. The first of those HBSs exploits only a single degree of freedom of the extended txn, namely the time investment, but it allows already for txns with a variable level of trust together with aligned network fees and energy usage. In principle, the last HBS should scale to tens of thousands timely txns per second while preserving what the previous HBSs achieved.
We also propose a simple homotopy based transition mechanism which enables us to relatively safely and continuously introduce new HBSs into the existing blockchain.
Our approach is constructive and as rigorous as possible and we attempt to analyze all aspects of these developments, al least at a conceptual level. The process is supported by evaluation on recent transaction data.

Replacements for Wed, 24 Apr 24

[8]  arXiv:2404.02029 (replaced) [pdf, other]

Title: Enhancing Portfolio Optimization with Transformer-GAN Integration: A Novel Approach in the Black-Litterman Framework

Authors: Enmin Zhu, Jerome Yen

Subjects: Computational Engineering, Finance, and Science (cs.CE)

[9]  arXiv:2404.12317 (replaced) [pdf, ps, other]

Title: Large Language Models for Synthetic Participatory Planning of Synergistic Transportation Systems

Authors: Jiangbo Yu

Subjects: Computational Engineering, Finance, and Science (cs.CE); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC); Multiagent Systems (cs.MA)

• New submissions
• Cross-lists
• Replacements