AI for Technoscientific Discovery: A Human-Inspired Architecture
Abstract
We present a high-level architecture for how artificial intelligences might advance and accumulate scientific and technological knowledge, inspired by emerging perspectives on how human intelligences advance and accumulate such knowledge. Agents advance knowledge by exercising a technoscientific method—an interacting combination of scientific and engineering methods. The technoscientific method maximizes a quantity we call “useful learning” via more-creative implausible utility (including the “aha!” moments of discovery), as well as via less-creative plausible utility. Society accumulates the knowledge advanced by agents so that other agents can incorporate and build on to make further advances. The proposed architecture is challenging but potentially complete: its execution might in principle enable artificial intelligences to advance and accumulate an equivalent of the full range of human scientific and technological knowledge.
- Authors:
- Publication Date:
- Research Org.:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Org.:
- USDOE National Nuclear Security Administration (NNSA); USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE Office of Science (SC)
- OSTI Identifier:
- 2311922
- Alternate Identifier(s):
- OSTI ID: 2311255
- Report Number(s):
- SAND-2024-01626J
Journal ID: ISSN 2713-3745; S2713374524000037; 100077; PII: S2713374524000037
- Grant/Contract Number:
- 230710; NA0003525
- Resource Type:
- Published Article
- Journal Name:
- Journal of Creativity (Online)
- Additional Journal Information:
- Journal Name: Journal of Creativity (Online) Journal Volume: 34 Journal Issue: 2; Journal ID: ISSN 2713-3745
- Publisher:
- Elsevier
- Country of Publication:
- United Kingdom
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; artificial intelligence; creativity; extended neurosymbolic knowledge; technoscientific method
Citation Formats
Tsao, J. Y., Abbott, R. G., Crowder, D. C., Desai, S., Dingreville, R. P. M., Fowler, J. E., Garland, A., Iyer, P. P., Murdock, J., Steinmetz, S. T., Yarritu, K. A., Johnson, C. M., and Stracuzzi, D. J. AI for Technoscientific Discovery: A Human-Inspired Architecture. United Kingdom: N. p., 2024.
Web. doi:10.1016/j.yjoc.2024.100077.
Tsao, J. Y., Abbott, R. G., Crowder, D. C., Desai, S., Dingreville, R. P. M., Fowler, J. E., Garland, A., Iyer, P. P., Murdock, J., Steinmetz, S. T., Yarritu, K. A., Johnson, C. M., & Stracuzzi, D. J. AI for Technoscientific Discovery: A Human-Inspired Architecture. United Kingdom. https://doi.org/10.1016/j.yjoc.2024.100077
Tsao, J. Y., Abbott, R. G., Crowder, D. C., Desai, S., Dingreville, R. P. M., Fowler, J. E., Garland, A., Iyer, P. P., Murdock, J., Steinmetz, S. T., Yarritu, K. A., Johnson, C. M., and Stracuzzi, D. J. Thu .
"AI for Technoscientific Discovery: A Human-Inspired Architecture". United Kingdom. https://doi.org/10.1016/j.yjoc.2024.100077.
@article{osti_2311922,
title = {AI for Technoscientific Discovery: A Human-Inspired Architecture},
author = {Tsao, J. Y. and Abbott, R. G. and Crowder, D. C. and Desai, S. and Dingreville, R. P. M. and Fowler, J. E. and Garland, A. and Iyer, P. P. and Murdock, J. and Steinmetz, S. T. and Yarritu, K. A. and Johnson, C. M. and Stracuzzi, D. J.},
abstractNote = {We present a high-level architecture for how artificial intelligences might advance and accumulate scientific and technological knowledge, inspired by emerging perspectives on how human intelligences advance and accumulate such knowledge. Agents advance knowledge by exercising a technoscientific method—an interacting combination of scientific and engineering methods. The technoscientific method maximizes a quantity we call “useful learning” via more-creative implausible utility (including the “aha!” moments of discovery), as well as via less-creative plausible utility. Society accumulates the knowledge advanced by agents so that other agents can incorporate and build on to make further advances. The proposed architecture is challenging but potentially complete: its execution might in principle enable artificial intelligences to advance and accumulate an equivalent of the full range of human scientific and technological knowledge.},
doi = {10.1016/j.yjoc.2024.100077},
journal = {Journal of Creativity (Online)},
number = 2,
volume = 34,
place = {United Kingdom},
year = {Thu Aug 01 00:00:00 EDT 2024},
month = {Thu Aug 01 00:00:00 EDT 2024}
}
https://doi.org/10.1016/j.yjoc.2024.100077
Works referenced in this record:
Generalizing from a Few Examples
journal, June 2020
- Wang, Yaqing; Yao, Quanming; Kwok, James T.
- ACM Computing Surveys, Vol. 53, Issue 3
A History of Microwave Heating Applications
journal, September 1984
- Osepchuk, J. M.
- IEEE Transactions on Microwave Theory and Techniques, Vol. 32, Issue 9
Science: A ‘Dappled World’ or a ‘Seamless Web’?
journal, September 2001
- Anderson, Philip W.
- Studies in History and Philosophy of Science Part B: Studies in History and Philosophy of Modern Physics, Vol. 32, Issue 3
Polanyi's revenge and AI's new romance with tacit knowledge
journal, January 2021
- Kambhampati, Subbarao
- Communications of the ACM, Vol. 64, Issue 2
Explainable Machine Learning for Scientific Insights and Discoveries
journal, January 2020
- Roscher, Ribana; Bohn, Bastian; Duarte, Marco F.
- IEEE Access, Vol. 8
A historical survey of algorithms and hardware architectures for neural-inspired and neuromorphic computing applications
journal, January 2017
- James, Conrad D.; Aimone, James B.; Miner, Nadine E.
- Biologically Inspired Cognitive Architectures, Vol. 19
Continual lifelong learning with neural networks: A review
journal, May 2019
- Parisi, German I.; Kemker, Ronald; Part, Jose L.
- Neural Networks, Vol. 113
Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
journal, February 2019
- Raissi, M.; Perdikaris, P.; Karniadakis, G. E.
- Journal of Computational Physics, Vol. 378
Defining Creativity: Don't We Also Need to Define What Is Not Creative?
journal, January 2016
- Simonton, Dean Keith
- The Journal of Creative Behavior, Vol. 52, Issue 1
The collective intelligence of evolution and development
journal, April 2023
- Watson, Richard; Levin, Michael
- Collective Intelligence, Vol. 2, Issue 2
Technological paradigms and technological trajectories
journal, June 1982
- Dosi, Giovanni
- Research Policy, Vol. 11, Issue 3
Inverse design in search of materials with target functionalities
journal, March 2018
- Zunger, Alex
- Nature Reviews Chemistry, Vol. 2, Issue 4
Autonomous experimentation systems for materials development: A community perspective
journal, September 2021
- Stach, Eric; DeCost, Brian; Kusne, A. Gilad
- Matter, Vol. 4, Issue 9
Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery
journal, May 2021
- Rosen, Andrew S.; Iyer, Shaelyn M.; Ray, Debmalya
- Matter, Vol. 4, Issue 5
Autonomous materials discovery and manufacturing (AMDM): A review and perspectives
journal, August 2022
- Bukkapatnam, Satish T. S.
- IISE Transactions, Vol. 55, Issue 1
A Dynamic Network Measure of Technological Change
journal, March 2017
- Funk, Russell J.; Owen-Smith, Jason
- Management Science, Vol. 63, Issue 3
What is missing in autonomous discovery: open challenges for the community
journal, January 2023
- Maffettone, Phillip M.; Friederich, Pascal; Baird, Sterling G.
- Digital Discovery, Vol. 2, Issue 6
Modular exaptation: A missing link in the synthesis of artificial form
journal, November 2014
- Andriani, Pierpaolo; Carignani, Giuseppe
- Research Policy, Vol. 43, Issue 9
Deep learning for AI
journal, July 2021
- Bengio, Yoshua; Lecun, Yann; Hinton, Geoffrey
- Communications of the ACM, Vol. 64, Issue 7
How to build an effective self-driving laboratory
journal, February 2023
- MacLeod, Benjamin P.; Parlane, Fraser G. L.; Berlinguette, Curtis P.
- MRS Bulletin, Vol. 48, Issue 2
Autonomous discovery in the chemical sciences part II: Outlook
journal, September 2019
- Coley, Connor W.; Eyke, Natalie S.; Jensen, Klavs F.
- Angewandte Chemie International Edition
The Robot Scientist Adam
journal, August 2009
- King, Ross D.; Rowland, Jem; Aubrey, Wayne
- Computer, Vol. 42, Issue 7
Autonomous (AI-driven) materials science
journal, September 2022
- Green, Martin L.; Maruyama, Benji; Schrier, Joshua
- Applied Physics Reviews, Vol. 9, Issue 3
Scientific Thinking in Young Children: Theoretical Advances, Empirical Research, and Policy Implications
journal, September 2012
- Gopnik, Alison
- Science, Vol. 337, Issue 6102
Autonomous discovery in the chemical sciences part I: Progress
journal, September 2019
- Jensen, Klavs F.; Coley, Connor W.; Eyke, Natalie S.
- Angewandte Chemie International Edition
On scientific understanding with artificial intelligence
journal, October 2022
- Krenn, Mario; Pollice, Robert; Guo, Si Yue
- Nature Reviews Physics, Vol. 4, Issue 12
Data-Driven Strategies for Accelerated Materials Design
journal, February 2021
- Pollice, Robert; dos Passos Gomes, Gabriel; Aldeghi, Matteo
- Accounts of Chemical Research, Vol. 54, Issue 4
The Blue LED Nobel Prize: Historical context, current scientific understanding, human benefit: The Blue LED Nobel Prize: Historical context, current scientific understanding, human benefit
journal, June 2015
- Tsao, Jeffrey Y.; Han, Jung; Haitz, Roland H.
- Annalen der Physik, Vol. 527, Issue 5-6
Creative Outcome as Implausible Utility
journal, July 2019
- Tsao, J. Y.; Ting, C. L.; Johnson, C. M.
- Review of General Psychology, Vol. 23, Issue 3
What do we know about the disruption index in scientometrics? An overview of the literature
journal, November 2023
- Leibel, Christian; Bornmann, Lutz
- Scientometrics, Vol. 129, Issue 1
The originality of machines: AI takes the Torrance Test
journal, December 2023
- Guzik, Erik E.; Byrge, Christian; Gilde, Christian
- Journal of Creativity, Vol. 33, Issue 3
Scientific discovery in the age of artificial intelligence
journal, August 2023
- Wang, Hanchen; Fu, Tianfan; Du, Yuanqi
- Nature, Vol. 620, Issue 7972