The 2024 Nobel Prize in Physics went to John Hopfield and Geoffrey Hinton for their basic research in artificial intelligence (AI), and the Nobel Prize in Chemistry went to David Baker and Demis for solving problems using AI. It went to Demis Hassabis and John Jumper. The protein folding problem has been a huge challenge in science for 50 years.
A new article written by researchers at Carnegie Mellon University and Calculation Consulting examines the convergence of physics, chemistry, and AI highlighted in the recent Nobel Prize. Traces the historical development of neural networks, emphasizing the role of interdisciplinary research in the advancement of AI. The authors advocate fostering AI-based polymaths to bridge the gap between theoretical developments and practical applications and promote progress toward artificial general intelligence. The article appears next: pattern.
“As AI is recognized in relation to both physics and chemistry, machine learning practitioners will need to understand how these sciences relate to AI and how these awards can impact their own work,” said Ganesh Mani, Carnegie Professor of Innovation Practice and Director of Collaborative AI. “You might wonder if it exists,” he explained. Mellon’s Tepper School of Business co-authored this article. “As we move forward, it is important to recognize the convergence of different approaches in forming modern AI systems based on generative AI.”
In their article, the authors explore the historical development of neural networks. They argue that by examining the history of AI development, we can more thoroughly understand the connections between computer science, theoretical chemistry, theoretical physics, and applied mathematics. A historical perspective illuminates how fundamental discoveries and inventions across these fields made modern machine learning through artificial neural networks possible.
We then look at the major innovations and challenges in the field, starting with Hopfield’s work, and, like the work of Jumper and Hassabis, explain how engineering sometimes outpaced scientific understanding.
The authors conclude with a call to action, suggesting that the rapid advancement of AI across a variety of fields presents both unprecedented opportunities and significant challenges. They say a new generation of interdisciplinary thinkers must be nurtured to bridge the gap between hype and real progress.
These “modern-day Leonardo da Vinci,” as the authors call them, will be critical to developing practical learning theories that engineers can immediately apply and to propel the field toward the ambitious goal of artificial general intelligence.
This, the authors say, requires a paradigm shift in how we approach scientific inquiry and problem solving: embracing holistic, interdisciplinary collaboration and learning from nature to understand it. By breaking down silos between disciplines and fostering a culture of intellectual curiosity across domains, we can identify innovative solutions to complex global challenges like climate change. The synthesis of diverse knowledge and perspectives facilitated by AI can enable meaningful progress and enable the field to realize the full potential of its technological aspirations.
“This interdisciplinary approach is not only beneficial, it is essential to solving many of the complex problems ahead,” suggests Charles Martin, principal consultant at Calculation Consulting, who co-authored the article. “We must capitalize on the momentum of current developments while remaining grounded in practical realities.”
The authors acknowledge the contribution of Scott E. Fahlman, Professor Emeritus of Computer Science at Carnegie Mellon.
