A lot has changed within 15 years from Kaim that he was a PhD student.
“When you are at the stage of doctoral students, there is a high wall between various disciplines and entities, and there was even a high wall in computer science,” he says. “A guy sitting next to me can do things that I couldn't understand.”
Within seven months, since he joined the myth of Schwarzman College of Computing as a professor of career development Douglas Ross (1954) software technology in the Department of Electrical Engineering and Computer Science, he says that he experiences something that in his opinion is “very rare in human scientific history” – lowering the walls that expand on various scientific disciplines.
“There is no way to understand high energy physics, chemistry or biology research limit, but now we see something that can help us break these walls,” he says, “And this is creating a common language that was found in AI.”
Building AI bridge
According to he, this change began in 2012 as a result of a “deep learning revolution”, in which it was aware that this set of machine learning methods based on neural networks was so powerful that it could be used.
“At this point, a computer vision – helping computers to see and perceive the world as if they were human beings – it began to grow very quickly, because, as it turns out, you can use the same methodology for many different problems and many different areas,” he says. “So the computer vision community quickly became really big, because these various subtopics were now able to speak a common language and share a common set of tools.”
From there, he says that the trend began to expand to other areas of computer science, including natural language processing, speech and robotics recognition, creating the basics of chatgpt and other progress towards artificial general intelligence (Aga).
“All this happened over the past decade, leading us to a new, emerging trend that I really can't wait for, and it observes how the AI methodology promotes other scientific disciplines,” he says.
He says that one of the most famous examples is Alphafold, an artificial intelligence program developed by Google Deepmind, which performs forecasts of protein structure.
“This is a completely different scientific discipline, a completely different problem, but people also use the same set of AI tools, the same methodology to solve these problems,” he says, “and I think this is just the beginning.”
The future of artificial intelligence in science
Since the arrival in the myth in February 2024, he says that he talked to professors in almost every faculty. On some days he talks to two or more professors from very different environments.
“Certainly I do not quite understand their research area, but they will simply introduce a certain context, and then we can start talking about deep learning, machine learning (i) neural networks models in their problems,” he says. “In this sense, these AI tools are like a common language between these scientific areas: machine learning tools” translate “their terminology and concepts into conditions that I can understand, and then I can learn their problems and share my experience, and sometimes propose solutions or opportunities to discover.”
The extension to various scientific disciplines has significant potential, from the use of video analysis to predict weather and climate trends to acceleration of the research cycle and reducing costs in relation to new drug discoveries.
While AI tools ensure a clear benefit of his colleagues scientists, he also notices the mutual effect that they can have and had, the creation and development of AI.
“Scientists are new problems and challenges that help us in the development of these tools,” he says. “But it is also important to remember that many of today's AI tools result from earlier scientific areas – for example, artificial neural networks have been inspired by biological observations; diffusion models to generate image were motivated from the term physics.”
“Science and AI are not isolated people. We are approaching the same goal from different perspectives, and now we meet.”
And is it a better place where they could meet than a myth.
“No wonder that the myth can see this change earlier than in many other places,” he says. “(Myth Schwarzman College of Computing) created an environment that connects different people and allows them to sit together, talk together, cooperate, exchange their ideas, talking about the same language – and I see that it starts.”
As for when the walls are fully reduced, it notes that it is a long -term investment that will not take place overnight.
“Several dozen years ago, computers were considered advanced technologies and you needed specific knowledge to understand them, but now everyone is using a computer,” he says. “I expect for 10 or more years, everyone will somehow use some artificial intelligence for their research – these are just their basic tools, their basic language and can use artificial intelligence to solve their problems.”
