Simulation Intelligence: Towards a New Generation of Scientific Methods

Lavin, Alexander; Zenil, Héctor; Krakauer, David C.; Gottschlich, Justin; Mattson, Tim; Anandkumar, Anima; Choudry, Sanjay; Rocki, Kamil; Baydin, Atılım Güneş; Prunkl, Carina; Paige, Brooks; Isayev, Olexandr; Peterson, Erik; McMahon, Peter L.; Macke, Jakob H.; Cranmer, K.; Zhang, Jiaxin; Wainwright, H. M.; Hanuka, Adi; Veloso, Manuela; Assefa, Samuel; Zheng, Stephan; Pfeffer, Avi

doi:10.48550/arxiv.2112.03235

Cited by 25 publications

(43 citation statements)

References 374 publications

(495 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This involves moving away from using AI to simply identify local material extrema and to start imbuing them with mechanistic and heuristic rules and tacit knowledge so that we can increase our understanding of why some materials are exceptional and others are mediocre. 10 The next step requires transcending barriers between disciplines and identifying challenges that present intellectually interesting opportunities across multiple disciplines including materials science, computer science, mechatronics, and general data science.…”

Section: Materials Acceleration Platformsmentioning

confidence: 99%

Reaching critical MASS: Crowdsourcing designs for the next generation of materials acceleration platforms

Seifrid

Hattrick‐Simpers

Aspuru‐Guzik

et al. 2022

Matter

View full text Add to dashboard Cite

Section: Materials Acceleration Platformsmentioning

confidence: 99%

Reaching critical MASS: Crowdsourcing designs for the next generation of materials acceleration platforms

Seifrid

Hattrick‐Simpers

Aspuru‐Guzik

et al. 2022

Matter

View full text Add to dashboard Cite

“…Note that this method does not impose any explicit assumption on the noise distribution (see equation 3), and the information about the forward model is implicitly encoded in the model and data pairs. As a result, this formulation is an instance of likelihood-free simulation-based inference methods [65,66] that allows us to approximate the posterior distribution for previously unseen data as,…”

Section: Amortized Variational Inferencementioning

confidence: 99%

Reliable amortized variational inference with physics-based latent distribution correction

Siahkoohi¹,

Rizzuti²,

Orozco³

et al. 2022

Preprint

View full text Add to dashboard Cite

Bayesian inference for high-dimensional inverse problems is challenged by the computational costs of the forward operator and the selection of an appropriate prior distribution. Amortized variational inference addresses these challenges where a neural network is trained to approximate the posterior distribution over existing pairs of model and data. When fed previously unseen data and normally distributed latent samples as input, the pretrained deep neural network-in our case a conditional normalizing flow-provides posterior samples with virtually no cost. However, the accuracy of this approach relies on the availability of high-fidelity training data, which seldom exists in geophysical inverse problems because of the highly heterogeneous structure of the Earth. In addition, accurate amortized variational inference requires the observed data to be drawn from the training data distribution. As such, we offer a solution that increases the resilience of amortized variational inference when faced with data distribution shifts, e.g., changes in the forward model or prior distribution. Our method involves a physics-based correction to the conditional normalizing flow latent distribution to provide a more accurate approximation to the posterior distribution for the observed data at hand. To accomplish this, instead of a standard Gaussian latent distribution, we parameterize the latent distribution by a Gaussian distribution with an unknown mean and diagonal covariance. These unknown quantities are then estimated by minimizing the Kullback-Leibler divergence between the corrected and true posterior distributions. While generic and applicable to other inverse problems, by means of a seismic imaging example, we show that our correction step improves the robustness of amortized variational inference with respect to changes in number of source experiments, noise variance, and shifts in the prior distribution. This approach provides a seismic image with limited artifacts and an assessment of its uncertainty with approximately the same cost as five reverse-time migrations.

show abstract

“…This requires constraining the ML techniques to produce results that are compatible to the real world. These methodological choices can either be in the form of transformations and feature selection to address observational biases, architectural choices (inductive biases—based on physical laws, boundary conditions) or other implicit learning biases (careful architectural choices to indirectly enforce physical constraints) [ 138 ].…”

Section: Multi-resolution Gaussian Processesmentioning

confidence: 99%

Machine Learning Methods for Multiscale Physics and Urban Engineering Problems

Sharma

Thompson

Laefer

et al. 2022

Entropy

View full text Add to dashboard Cite

We present an overview of four challenging research areas in multiscale physics and engineering as well as four data science topics that may be developed for addressing these challenges. We focus on multiscale spatiotemporal problems in light of the importance of understanding the accompanying scientific processes and engineering ideas, where “multiscale” refers to concurrent, non-trivial and coupled models over scales separated by orders of magnitude in either space, time, energy, momenta, or any other relevant parameter. Specifically, we consider problems where the data may be obtained at various resolutions; analyzing such data and constructing coupled models led to open research questions in various applications of data science. Numeric studies are reported for one of the data science techniques discussed here for illustration, namely, on approximate Bayesian computations.

show abstract

Simulation Intelligence: Towards a New Generation of Scientific Methods

Cited by 25 publications

References 374 publications

Reaching critical MASS: Crowdsourcing designs for the next generation of materials acceleration platforms

Reaching critical MASS: Crowdsourcing designs for the next generation of materials acceleration platforms

Reliable amortized variational inference with physics-based latent distribution correction

Machine Learning Methods for Multiscale Physics and Urban Engineering Problems

Contact Info

Product

Resources

About