ZME Science
No Result
View All Result
ZME Science
No Result
View All Result
ZME Science

Home → Science

Researchers teach AI to design, say it did ‘quite good’ but won’t steal your job (yet)

The AI is "learning how people solve a specific type of problem and creating new design solutions from scratch."

Alexandru MicubyAlexandru Micu
November 7, 2019
in News, Science, Tech, Technology
A A
Share on FacebookShare on TwitterSubmit to Reddit

A US-based research team has trained artificial intelligence (AI) in design, with pretty good results.

A roof supported by a wooden truss framework.
Image credits Achim Scholty.

Although we don’t generally think of AIs as good problem-solvers, a new study suggests they can learn how to be. The paper describes the process through which a framework of deep neural networks learned human creative processes and strategies and how to apply them to create new designs.

Just hit ‘design’

“We were trying to have the [AIs] create designs similar to how humans do it, imitating the process they use: how they look at the design, how they take the next action, and then create a new design, step by step,” says Ayush Raina, a Ph.D. candidate in mechanical engineering at Carnegie Mellon and a co-author of the study.

Design isn’t an exact science. While there are definite no-no’s and rules of thumb that lead to OK designs, good designs require creativity and exploratory decision-making. Humans excel at these skills.

Software as we know it today works wonders within a clearly defined set of rules, with clear inputs and known desired outcomes. That’s very handy when you need to crunch huge amounts of data, or to make split-second decisions to keep a jet stable in flight, for example. However, it’s an appalling skillset for someone trying their hand, or processors, at designing.

The team wanted to see if machines can learn the skills that make humans good designers and then apply them. For the study, they created an AI framework from several deep neural networks and fed it data pertaining to a human going about the process of design.

The study focused on trusses, which are complex but relatively common design challenges for engineers. Trusses are load-bearing structural elements composed of rods and beams; bridges and large buildings make good use of trusses, for example. Simple in theory, trusses are actually incredibly complex elements whose final shapes are a product of their function, material make-up, or other desired traits (such as flexibility-rigidity, resistance to compression-tension and so forth).

The framework itself was made up of several deep neural networks which worked together in a prediction-based process. It was shown five successive snapshots of the structures (the design modification sequence for a truss), and then asked to predict the next iteration of the design. This data was the same one engineers use when approaching the problem: pixels on a screen; however, the AI wasn’t privy to any further information or context (such as the truss’ intended use). The researchers emphasized visualization in the process because vision is an integral part of how humans perceive the world and go about solving problems.

RelatedPosts

Solar Hourglass might power 1,000 Danish Homes while inspiring Climate Change Action
The NSA is gathering so much data, it’s become swamped and ironically ineffective at preventing terrorism
Trippy AI writes interactive text adventure game on the fly
Scientists brew the future with AI that ‘tastes’ Belgian beer to make it better

In essence, the researchers had their neural networks watch human designers throughout the whole design process, and then try to emulate them. Overall, the team reports, the way their AI approached the design process was similar to that employed by humans. Further testing on similar design problems showed that on average, the AI can perform just as well if not better than humans. However, the system still lacks many of the advantages a human user would have when problem-solving — namely, it worked without a specific goal in mind (a particular weight or shape, for example), and didn’t receive feedback on how successful it was on its task. In other words, while the program could design a good truss, it didn’t understand what it was doing, what the end goal of the process was, or how good it was at it. So while it’s good at designing, it’s still a lousy designer.

All things considered, however, the AI was “quite good” at the task, says co-author Jonathan Cagan, professor of mechanical engineering and interim dean of Carnegie Mellon University’s College of Engineering.

“The AI is not just mimicking or regurgitating solutions that already exist,” Professor Cagan explains. “It’s learning how people solve a specific type of problem and creating new design solutions from scratch.”

“It’s tempting to think that this AI will replace engineers, but that’s simply not true,” said Chris McComb, an assistant professor of engineering design at the Pennsylvania State University and paper co-author.

“Instead, it can fundamentally change how engineers work. If we can offload boring, time-consuming tasks to an AI, like we did in the work, then we free engineers up to think big and solve problems creatively.”

The paper “Learning to Design From Humans: Imitating Human Designers Through Deep Learning” has been published in the Journal of Mechanical Design.

Tags: AIArtificialdesignintelligencelearningmachineTruss

ShareTweetShare
Alexandru Micu

Alexandru Micu

Stunningly charming pun connoisseur, I have been fascinated by the world around me since I first laid eyes on it. Always curious, I'm just having a little fun with some very serious science.

Related Posts

Future

GPT-5 is, uhm, not what we expected. Has AI just plateaued?

byMichael Rovatsos
2 days ago
Health

AI Can Hear Cancer in the Voice Before Doctors Can Detect It

byMihai Andrei
5 days ago
Future

Illinois Just Became the First State to Ban AI From Acting as a Therapist

byTudor Tarita
1 week ago
a robot sitting with "evil" writing on its arm
Future

Anthropic says it’s “vaccinating” its AI with evil data to make it less evil

byMihai Andrei
2 weeks ago

Recent news

The UK Government Says You Should Delete Emails to Save Water. That’s Dumb — and Hypocritical

August 16, 2025

In Denmark, a Vaccine Is Eliminating a Type of Cervical Cancer

August 16, 2025
This Picture of the Week shows a stunning spiral galaxy known as NGC 4945. This little corner of space, near the constellation of Centaurus and over 12 million light-years away, may seem peaceful at first — but NGC 4945 is locked in a violent struggle. At the very centre of nearly every galaxy is a supermassive black hole. Some, like the one at the centre of our own Milky Way, aren’t particularly hungry. But NGC 4945’s supermassive black hole is ravenous, consuming huge amounts of matter — and the MUSE instrument at ESO’s Very Large Telescope (VLT) has caught it playing with its food. This messy eater, contrary to a black hole’s typical all-consuming reputation, is blowing out powerful winds of material. This cone-shaped wind is shown in red in the inset, overlaid on a wider image captured with the MPG/ESO telescope at La Silla. In fact, this wind is moving so fast that it will end up escaping the galaxy altogether, lost to the void of intergalactic space. This is part of a new study that measured how winds move in several nearby galaxies. The MUSE observations show that these incredibly fast winds demonstrate a strange behaviour: they actually speed up far away from the central black hole, accelerating even more on their journey to the galactic outskirts. This process ejects potential star-forming material from a galaxy, suggesting that black holes control the fates of their host galaxies by dampening the stellar birth rate. It also shows that the more powerful black holes impede their own growth by removing the gas and dust they feed on, driving the whole system closer towards a sort of galactic equilibrium. Now, with these new results, we are one step closer to understanding the acceleration mechanism of the winds responsible for shaping the evolution of galaxies, and the history of the universe. Links  Research paper in Nature Astronomy by Marconcini et al. Close-up view of NGC 4945’s nucleus

Astronomers Find ‘Punctum,’ a Bizarre Space Object That Might be Unlike Anything in the Universe

August 15, 2025
  • About
  • Advertise
  • Editorial Policy
  • Privacy Policy and Terms of Use
  • How we review products
  • Contact

© 2007-2025 ZME Science - Not exactly rocket science. All Rights Reserved.

No Result
View All Result
  • Science News
  • Environment
  • Health
  • Space
  • Future
  • Features
    • Natural Sciences
    • Physics
      • Matter and Energy
      • Quantum Mechanics
      • Thermodynamics
    • Chemistry
      • Periodic Table
      • Applied Chemistry
      • Materials
      • Physical Chemistry
    • Biology
      • Anatomy
      • Biochemistry
      • Ecology
      • Genetics
      • Microbiology
      • Plants and Fungi
    • Geology and Paleontology
      • Planet Earth
      • Earth Dynamics
      • Rocks and Minerals
      • Volcanoes
      • Dinosaurs
      • Fossils
    • Animals
      • Mammals
      • Birds
      • Fish
      • Amphibians
      • Reptiles
      • Invertebrates
      • Pets
      • Conservation
      • Animal facts
    • Climate and Weather
      • Climate change
      • Weather and atmosphere
    • Health
      • Drugs
      • Diseases and Conditions
      • Human Body
      • Mind and Brain
      • Food and Nutrition
      • Wellness
    • History and Humanities
      • Anthropology
      • Archaeology
      • History
      • Economics
      • People
      • Sociology
    • Space & Astronomy
      • The Solar System
      • Sun
      • The Moon
      • Planets
      • Asteroids, meteors & comets
      • Astronomy
      • Astrophysics
      • Cosmology
      • Exoplanets & Alien Life
      • Spaceflight and Exploration
    • Technology
      • Computer Science & IT
      • Engineering
      • Inventions
      • Sustainability
      • Renewable Energy
      • Green Living
    • Culture
    • Resources
  • Videos
  • Reviews
  • About Us
    • About
    • The Team
    • Advertise
    • Contribute
    • Editorial policy
    • Privacy Policy
    • Contact

© 2007-2025 ZME Science - Not exactly rocket science. All Rights Reserved.