Artificial Intelligence in Physical Sciences: Symbolic Regression Trends and Perspectives

Angelis, Dimitrios; Sofos, Filippos; Karakasidis, Theodoros E.

doi:10.1007/s11831-023-09922-z

Cited by 34 publications

(20 citation statements)

References 228 publications

(228 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, symbolic regression hybridized with genetic programming has been shown to provide analytical equations from data. 176…”

Section: ■ Historical Tools (<1990)mentioning

confidence: 99%

“…Also, before the possibility for accurate quantitative descriptions of the various fluidization phenomena, PINN promises to supersede the current empirical correlations that have been known to make predictions that are different orders of magnitude (e.g., entrainment). In particular, symbolic regression hybridized with genetic programming has been shown to provide analytical equations from data …”

Section: Tomorrow’s Tools (>2020)mentioning

confidence: 99%

See 1 more Smart Citation

Fluidized Bed Scale-Up for Sustainability Challenges. 1. Tomorrow’s Tools

Cocco,

Chew

2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

The scaling up of fluidized beds has been purposefully pursued for more than 100 years. Yet, over that time, scale-up tools have not significantly changed. Data analysis is typically a standard analysis of variances statistical exercise, perhaps reinforced with a design of experimental procedure. Flowsheeting and equipment design are based on institutional knowledge, albeit graphical user interface-based process flow models make that job more manageable. Advanced models such as computational fluid dynamics are used but often as a supplement and not a primary driver. As a result, the scale-up process for a fluidized bed can take more than 10 years. Fluidized beds remain at the forefront of the present time-critical sustainability challenges, e.g., carbon capture by particulate sorbents, methane-to-hydrogen, plastic-to-chemicals, etc. In view of the exigency toward net zero, today's scale-up efforts need to be accelerated, leveraging the advanced new tools that have become readily available. The problem is that such tools are often neglected, inadequately implemented, ineffectively resourced, and/or poorly understood. This motivated the current effort, which is targeted at reviewing how scale-up tools have evolved over the years and the promising new tools, addressing some of the barriers of these tools in the design and scale-up of fluidized beds, as well as contemplating what can be done to circumvent these barriers. As a follow up, a companion part 2 (Cocco, R. A.; Chew, J. W. Ind. Eng. Chem. Res., submitted for publication) proposes a new scale-up path leveraging the advanced tools to achieve timely implementation of the new green fluidized bed processes.

show abstract

“…In particular, symbolic regression hybridized with genetic programming has been shown to provide analytical equations from data. 176…”

Section: ■ Historical Tools (<1990)mentioning

confidence: 99%

Section: Tomorrow’s Tools (>2020)mentioning

confidence: 99%

Fluidized Bed Scale-Up for Sustainability Challenges. 1. Tomorrow’s Tools

Cocco,

Chew

2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…This approach holds promise for analyzing high-throughput CRISPR-Cas-based knockout experiments, offering valuable insights for plant breeding ( Van Huffel et al., 2022 ). Symbolic regression uses genetic programming to automatically discover a white-box model of one system ( Angelis et al., 2023 ; Cranmer, 2023 ) – ideal to find the earlier-discussed meta-mechanisms. The DreamCoder system can uncover simple programs that generate example datasets ( Ellis et al., 2023 ).…”

Section: Nine Simulation Intelligence Motifs For Plant Sciencementioning

confidence: 99%

Plant science in the age of simulation intelligence

Stock,

Pieters,

De Swaef

et al. 2024

Front. Plant Sci.

View full text Add to dashboard Cite

Historically, plant and crop sciences have been quantitative fields that intensively use measurements and modeling. Traditionally, researchers choose between two dominant modeling approaches: mechanistic plant growth models or data-driven, statistical methodologies. At the intersection of both paradigms, a novel approach referred to as “simulation intelligence”, has emerged as a powerful tool for comprehending and controlling complex systems, including plants and crops. This work explores the transformative potential for the plant science community of the nine simulation intelligence motifs, from understanding molecular plant processes to optimizing greenhouse control. Many of these concepts, such as surrogate models and agent-based modeling, have gained prominence in plant and crop sciences. In contrast, some motifs, such as open-ended optimization or program synthesis, still need to be explored further. The motifs of simulation intelligence can potentially revolutionize breeding and precision farming towards more sustainable food production.

show abstract

“…Multiple methods for incorporating neural networks into SR have been developed, ranging from powerful problem simplification schemes (Cranmer et al 2020;Udrescu & Tegmark 2020;Udrescu et al 2020), to end-to-end SR methods where a neural network is trained in a supervised manner to map the relationship between data sets and their corresponding symbolic functions (Biggio et al 2020;Aréchiga et al 2021;Biggio et al 2021;Alnuqaydan et al 2022;Becker et al 2022;d'Ascoli et al 2022;Vastl et al 2022;Bendinelli et al 2023;Kamienny et al 2023), all the way to incorporating symbols into neural networks and sparsely fitting them to enable interpretability or to recover a mathematical expression (Brunton et al 2016;Martius & Lampert 2017;Ouyang et al 2018;Sahoo et al 2018;Kim et al 2020;Panju & Ghodsi 2020;Valle & Haddadin 2021;Zheng et al 2022). See La Cava et al (2021), Makke & Chawla (2022), and Angelis et al (2023) for recent reviews of SR algorithms.…”

Section: Related Work-a Brief Survey Of Modern Srmentioning

confidence: 99%

Deep Symbolic Regression for Physics Guided by Units Constraints: Toward the Automated Discovery of Physical Laws

Tenachi,

Ibata,

Diakogiannis

2023

ApJ

View full text Add to dashboard Cite

Symbolic regression (SR) is the study of algorithms that automate the search for analytic expressions that fit data. While recent advances in deep learning have generated renewed interest in such approaches, the development of SR methods has not been focused on physics, where we have important additional constraints due to the units associated with our data. Here we present Φ-SO, a physical symbolic optimization framework for recovering analytical symbolic expressions from physics data using deep reinforcement learning techniques by learning units constraints. Our system is built, from the ground up, to propose solutions where the physical units are consistent by construction. This is useful not only in eliminating physically impossible solutions but also because the grammatical rules of dimensional analysis enormously restrict the freedom of the equation generator, thus vastly improving performance. The algorithm can be used to fit noiseless data, which can be useful, for instance, when attempting to derive an analytical property of a physical model, and it can also be used to obtain analytical approximations of noisy data. We test our machinery on a standard benchmark of equations from the Feynman Lectures on Physics and other physics textbooks, achieving state-of-the-art performance in the presence of noise (exceeding 0.1%) and show that it is robust even in the presence of substantial (10%) noise. We showcase its abilities on a panel of examples from astrophysics.

show abstract

Artificial Intelligence in Physical Sciences: Symbolic Regression Trends and Perspectives

Cited by 34 publications

References 228 publications

Fluidized Bed Scale-Up for Sustainability Challenges. 1. Tomorrow’s Tools

Fluidized Bed Scale-Up for Sustainability Challenges. 1. Tomorrow’s Tools

Plant science in the age of simulation intelligence

Deep Symbolic Regression for Physics Guided by Units Constraints: Toward the Automated Discovery of Physical Laws

Contact Info

Product

Resources

About