Theory-Guided Data Science: A New Paradigm for Scientific Discovery from Data
By Anuj Karpatne, Gowtham Atluri, James H. Faghmous, Michael S. Steinbach, Arindam Banerjee, Auroop Ratan Ganguly, Shashi Shekhar, Nagiza F. Samatova, Vipin Kumar. 2016
1,347 citations recorded by the source corpus.
Data science models, although successful in a number of commercial domains, have had limited applicability in scientific problems involving complex physical phenomena. Theory-guided data science (TGDS) is an emerging paradigm that aims to leverage the wealth of scientific knowledge for improving the effectiveness of data science models in enabling scientific discovery. The overarching vision of TGDS is to introduce scientific consistency as an essential component for learning generalizable models. Further, by producing scientifically interpretable models, TGDS aims to advance our scientific understanding by discovering novel domain insights. Indeed, the paradigm of TGDS has started to gain prominence in a number of scientific disciplines such as turbulence modeling, material discovery, quantum chemistry, bio-medical science, bio-marker discovery, climate science, and hydrology. In this paper, we formally conceptualize the paradigm of TGDS and present a taxonomy of research themes in TGDS. We describe several approaches for integrating domain knowledge in different research themes using illustrative examples from different disciplines. We also highlight some of the promising avenues of novel research for realizing the full potential of theory-guided data science.
Subfields: Artificial Intelligence, Information Systems, Information Systems and Management, Numerical Methods & Scientific Computing, AI for Physical Sciences
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Francisco Herrera
- Chaopeng Shen
- Merouane Debbah
- Salvador Garcia
- Medhi Bennis
- Yoshua Bengio
- Percy Liang
- Sergey Levine
In Information Systems
- Francisco Herrera
- Salvador Garcia
- Vipin Kumar
- Michael S. Steinbach
- Tanya Y. Berger-Wolf
- Trevor Campbell
- Daniel Molina
- Walid Saad
In Information Systems and Management
- Michael S. Steinbach
- Xiaowei Jia
- Jared D. Willard
- Arvid Lundervold
- Francisco Herrera
- Cèsar Ferri
- Vipin Kumar
- Sepp Hochreiter
In Numerical Methods & Scientific Computing
- Adrian Albert
- Merouane Debbah
- Arvid Lundervold
- Christian Bauckhage
- Medhi Bennis
- Michael S. Steinbach
- Majid Ebad Sichani
- Yoshua Bengio
In AI for Physical Sciences
- Michael S. Steinbach
- Xiaowei Jia
- Jared D. Willard
- Adrian Albert
- Ehsan Haghighat
- Trevor Campbell
- Arvid Lundervold
- Majid Ebad Sichani
Citations in the corpus
- Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, positive): “All the additions presented in [374] push toward techniques that would eventually render a model explainable, and furthermore, knowledge consistent.”
- An overview of deep learning in medical imaging focusing on MRI (2018, background): “learning hybrids [12,13], formed to reduce computational costs by having the”
- Spatio-Temporal Data Mining (2017, methodology): “This is the paradigm of theory-guided data science [Karpatne et al. 2017] that is gaining attention in several disciplines for accelerating scientific discovery from data, and is particularly relevant for STDM applications.”
- Can deep learning beat numerical weather prediction? (2021, background): “The incorporation of physical laws in NNs is becoming a vibrant area of research (e.g. [125–128]) and is now often denoted as scientific ML.”
- Modeling process–structure–property relationships in metal additive manufacturing: a review on physics-driven versus data-driven approaches (2020, background): “A new paradigm of theory-guided data science was introduced by Karpatne et al [114], in which physics-driven knowledge and data science are integrated to leverage the potential of data science in the advancement of materials discovery.”
- Multi-Fidelity Physics-Constrained Neural Network and Its Application in Materials Modeling (2019, background): “The physical laws or models can guide the searching and optimization procedures [11].”
- Physics-guided Neural Networks (PGNN): An Application in Lake Temperature Modeling (2017, background): “Physics-guided data science (PGDS) is an emerging paradigm that aims to leverage the wealth of physical knowledge for improving the effectiveness of data science models in enabling scientific discovery [9].”
- Explainable Machine Learning for Scientific Insights and Discoveries (2019, background): “Reference [35] underlines the importance of consistency by defining it as an essential component to measure performance:”
- Digital twin, physics-based model, and machine learning applied to damage detection in structures (2020, methodology): “There are many possibilities for this integration, some of which are out of the scope of the present work, such as using machine learning to discover governing equations [32], using machine learning to solve partial differential equations [33], incorporating physical constraints into the machine lea”
- Integrating Physics-Based Modeling with Machine Learning: A Survey (2020, methodology): “Machine learning (ML) models, which have already found tremendous success in commercial applications, are beginning to play an important role in advancing scientific discovery in domains traditionally dominated by mechanistic (e.g. first principle) models [22, 112, 117, 126, 127, 138, 214, 261].”
- Physics-informed machine learning: case studies for weather and climate modelling (2021, background): “Karpatne et al. [11] and Willard et al. [12] broadly survey research work in this field that applies across the sciences.”
- Informed Machine Learning – A Taxonomy and Survey of Integrating Prior Knowledge into Learning Systems (2019, background): “theory-guided data science and points out the importance of enforcing scientific consistency inmachine learning [22].”