Blog Posts

• Take a look at our “Behind the Paper” Blog Post on Springer Nature Protocols and Methods Community
• Blog post “AI-Descartes: A framework for automated scientific discovery” on IBM-Research website.

Media coverage

Radio

• Radio interview on the “Quirks and Quarks” program on the CBC Radio Canada the Canadian public broadcaster.
• Radio interview on the “Deutschlandfunk Kultur” program of Deutschlandradio the national German public radio broadcaster for culture and science.

Press release and some articles

• “AI-Descartes: A Scientific Renaissance in the World of Artificial Intelligence”
• “What does it take to get AI to work like a scientist?”
• “‘AI scientist’ brings us a step closer to the age of machine-generated scientific discovery”
• “A NEW MACHINE-LEARNING TOOL THAT REASONS”
• “AI ‘scientist’ re-discovers scientific equations using data”
• “New “AI scientist” combines theory and data to discover scientific equations”

Talks & Videos

• Invited talk at the IBM Neuro-Symbolic AI Workshop (2023)
  • Presentation recording: Video
• Invited talk at the 2nd Workshop on Nobel Turing Challenge (2022)
  • More info on the Nobel Turing Challenge can be found HERE
  • Presentation recording: YouTube video
• Invited talk at the Paris Artificial Intelligence Research Institute: Colloquium PR[AI]RIE (2023)
  • Slides HERE

References

Papers:

• Combining data and theory for derivable scientific discovery with AI-Descartes, PDF C. Cornelio, S. Dash, V. Austel, T. Josephson, J. Goncalves, K. Clarkson, N. Megiddo, B. El Khadir, L. Horesh, Nature Communications, 2023
  • Featured on the Nature Editors’ Highlights webpage of recent research for “AI and machine learning”
  • Featured on the Nature Editors’ Highlights webpage of recent research for “Applied physics and mathematics”
• AI-Descartes: Integration of Data and Theory for Accelerated Derivable Symbolic Discovery, C. Cornelio, S. Dash, V. Austel, T. Josephson, J. Goncalves, K. Clarkson, N. Megiddo, B. El Khadir, L. Horesh, PrePrint arXiv:2109.01634, 2021
• Bayesian Experimental Design for Symbolic Discovery, K. L. Clarkson, C. Cornelio, S. Dash, J. Goncalves, L. Horesh, N. Megiddo, arXiv:2211.15860, 2022
• Symbolic Regression using Mixed-Integer Nonlinear Optimization, V. Austel, C. Cornelio, S. Dash, J. Goncalves, L. Horesh, T. Josephson, N. Megiddo, 2019, arXiv:2006.06813

Patents:

• Background Theory-Based Method for Refinement and Evaluation of Functional Models Extracted from Numerical Data, Lior Horesh, C. Cornelio, Bachir El Khadir, Sanjeeb Dash, Joao P. Goncalves, Kenneth Lee Clarkson, 2023
• Logical and Statistical Composite Models, L. Horesh, B. El Kadir, S. Dash, K. Clarkson, C. Cornelio, 2023
• Symbolic Model Discovery Rectification, L. Horesh, C. Cornelio, S. Dash, J.P. Goncalves, K. L. Clarkson, N. Megiddo, V. Austel, B. El Khadir, 2023
• Symbolic Model Discovery based on a combination of Numerical Learning Methods and Reasoning, C. Cornelio, L. Horesh, A. Fokoue-Nkoutche, S. Dash, 2020
• Generative Reasoning for Symbolic Discovery, C. Cornelio, L. Horesh, V. Pestun, R. Yan, 2020
• Experimental Design for Symbolic Model Discovery. L. Horesh, K. Clarkson, C. Cornelio, S. Magliacane, 2020