Toward Deep Generation of Guided Wave Representations for Composite Materials

Rautela, Mahindra; Senthilnath, J.; Huber, Armin; Gopalakrishnan, S.

doi:10.1109/tai.2022.3229653

Cited by 5 publications

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This distribution can be randomly sampled and processed through the trained decoder, enabling the generation of new realistic phase space projections across different modules. In comparison to Generative Adversarial Networks (GANs), VAEs, including CVAE, avoid issues like mode collapse and are generally easier to model and train more efficiently 64 . The ability to generate new data is crucial not only for increasing the dataset size for robust diagnostics but also for problems like system parameter estimation and control 24,65 .…”

Section: Generative Abilitymentioning

confidence: 99%

A conditional latent autoregressive recurrent model for generation and forecasting of beam dynamics in particle accelerators

Rautela,

Williams,

Scheinker

2024

Preprint

Self Cite

View full text Add to dashboard Cite

Particle accelerators are complex systems that focus, guide, and accelerate intense charged particle beams to high energy. Beam diagnostics present a challenging problem due to limited non-destructive measurements, computationally demanding simulations, and inherent uncertainties in the system. We propose a two-step unsupervised deep learning framework named as Conditional Latent Autoregressive Recurrent Model (CLARM) for learning the spatiotemporal dynamics of charged particles in accelerators. CLARM consists of a Conditional Variational Autoencoder (CVAE) transforming six-dimensional phase space into a lower-dimensional latent distribution and a Long Short-Term Memory (LSTM) network capturing temporal dynamics in an autoregressive manner. The CLARM can generate projections at various accelerator modules by sampling and decoding the latent space representation. The model also forecasts future states (downstream locations) of charged particles from past states (upstream locations). The results demonstrate that the generative and forecasting ability of the proposed approach is promising when tested against a variety of evaluation metrics.

show abstract

Section: Generative Abilitymentioning

confidence: 99%

A conditional latent autoregressive recurrent model for generation and forecasting of beam dynamics in particle accelerators

Rautela,

Williams,

Scheinker

2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We train the Convo-Implicit VAE to minimize the discrepancy between the reconstructed and input SDFs by reducing their mean squared error [24]. Additionally, the latent space (fig.…”

Section: The Latent Space Z and The Projected Coordinates C Are Conca...mentioning

confidence: 99%

PhoTOS: Topology Optimization of Photonic Components using a Shape Library

Padhy,

Chandrasekhar

2024

Preprint

View full text Add to dashboard Cite

Topology Optimization (TO) holds the promise of designing next-generation compact and efficient photonic components. However, ensuring the optimized designs comply with fabrication constraints imposed by semiconductor foundries remains a challenge. This work presents a TO framework that guarantees designs satisfy fabrication criteria, particularly minimum feature size and separation. Leveraging recent advancements in machine learning and feature mapping methods, our approach constructs components by transforming shapes from a predefined library, simplifying constraint enforcement. Specifically, we introduce a Convo-implicit Variational Autoencoder to encode the discrete shape library into a differentiable space, enabling gradient-based optimization. The efficacy of our framework is demonstrated through the design of several common photonic components.

show abstract

“…Note that the reconstruction will not be exact. The VAE is trained to minimize the difference between the output and input [66]. This involves optimizing the weights associated with the encoder and decoder.…”

Section: Variational Auto-encodersmentioning

confidence: 99%

“…Here, β is set to 10 −7 [66]. To achieve a stable convergence, the geometric parameters, contact area and volume fraction are normalized linearly between 0 and 1, while the permeability components are scaled logarithmically due to significant variation in magnitude.…”

Section: Variational Auto-encodersmentioning

confidence: 99%

TOMAS: Topology Optimization of Multiscale Fluid-Flow Devices using Variational Autoencoders and Super-Shapes

Padhy,

Suresh,

Chandrasekhar

2023

Preprint

View full text Add to dashboard Cite

In this paper, we present a framework for multiscale topology optimization of fluid-flow devices. The objective is to minimize dissipated power, subject to a desired contact-area. The proposed strategy is to design optimal microstructures in individual finite element cells, while simultaneously optimizing the overall fluid flow. In particular, parameterized super-shapes are chosen here to represent microstructures since they exhibit a wide range of permeability and contact area. To avoid repeated homogenization, a finite set of these super-shapes are analyzed a priori, and a variational autoencoder (VAE) is trained on their fluid constitutive properties (permeability), contact area and shape parameters. The resulting differentiable latent space is integrated with a coordinate neural network to carry out a global multi-scale fluid flow optimization. The latent space enables the use of new microstructures that were not present in the original data-set. The proposed method is illustrated using numerous examples in 2D Overview of the proposed method: A dataset of shape parameters of super-shapes, contact areas, and homogenized permeability tensors is used to train a variational autoencoder (VAE). The resulting latent space is then used for global multiscale optimization of fluid flow.

show abstract

Toward Deep Generation of Guided Wave Representations for Composite Materials

Cited by 5 publications

References 38 publications

A conditional latent autoregressive recurrent model for generation and forecasting of beam dynamics in particle accelerators

A conditional latent autoregressive recurrent model for generation and forecasting of beam dynamics in particle accelerators

PhoTOS: Topology Optimization of Photonic Components using a Shape Library

TOMAS: Topology Optimization of Multiscale Fluid-Flow Devices using Variational Autoencoders and Super-Shapes

Contact Info

Product

Resources

About