Slashdot Mirror
A Robot Scientist Will Dream Up New Materials To Advance Computing (technologyreview.com)
An anonymous reader quotes a report from MIT Technology Review: In a laboratory that overlooks a busy shopping street in Cambridge, Massachusetts, a robot is attempting to create new materials. A robot arm dips a pipette into a dish and transfers a tiny amount of bright liquid into one of many receptacles sitting in front of another machine. When all the samples are ready, the second machine tests their optical properties, and the results are fed to a computer that controls the arm. Software analyzes the results of these experiments, formulates a few hypotheses, and then starts the process over again. Humans are barely required.
The setup, developed by a startup called Kebotix, hints at how machine learning and robotic automation may be poised to revolutionize materials science in coming years. The company believes it may find new compounds that could, among other things, absorb pollution, combat drug-resistant fungal infections, and serve as more efficient optoelectronic components. The company's software learns from 3-D models of molecules with known properties. Kebotix uses several machine-learning methods to design novel chemical compounds. The company feeds molecular models of compounds with desirable properties into a type of neural network that learns a statistical representation of those properties. This algorithm can then come up with new examples that fit the same model. To strain out potentially useless materials, Kebotix uses another neural network and "then the company's robotic system tests the remaining chemical structures," reports MIT Technology Review. "The results of those experiments can be fed back into the machine-learning pipeline, helping it get closer to the desired chemical properties. The company dubs the overall system a 'self-driving lab.'"
The setup, developed by a startup called Kebotix, hints at how machine learning and robotic automation may be poised to revolutionize materials science in coming years. The company believes it may find new compounds that could, among other things, absorb pollution, combat drug-resistant fungal infections, and serve as more efficient optoelectronic components. The company's software learns from 3-D models of molecules with known properties. Kebotix uses several machine-learning methods to design novel chemical compounds. The company feeds molecular models of compounds with desirable properties into a type of neural network that learns a statistical representation of those properties. This algorithm can then come up with new examples that fit the same model. To strain out potentially useless materials, Kebotix uses another neural network and "then the company's robotic system tests the remaining chemical structures," reports MIT Technology Review. "The results of those experiments can be fed back into the machine-learning pipeline, helping it get closer to the desired chemical properties. The company dubs the overall system a 'self-driving lab.'"
Why does everything keep getting dumber and dumber and more obnoxious and broken if this "AI" is so great and will fix everything? Absolute, utter insanity.
As long as AI cannot distuigh a rifle from a turtle and gets all confused by an elephant that wasn't supposed to be there,
AI is only usefull for burning through your VC money.
But keep dreaming millennials. Try not to lose that when you reach the age that you realize Star Trek and The Matrix were something somebody made up.
As I always suspected, science is tedious and boring, work fit only for machines. I'm so glad I steered clear of STEW (Science Technology Engineering and Whatever).
The Trump traitors are safe, for now. Wait a minute or 4.
All I can come up with is showing up to an exam late in my underwear.
Sig. Sig. Sputnik
I wouldn't drink it if I were you.
Seriously, the only thing missing is blockchain.
Were supposed to inviolate IP laws surrounding the produced goods.
If an algorithm designs music without significant human creative input, or as in this case researches and produces results on a series of materials, the end result isn't supposed to be copyrightable or patentable since it wasn't a direct product of human innovation.
The corporations just keep getting more powerful while the little guy gets further and further from the feed.
The infinite monkey theorem states that a monkey hitting keys at random on a typewriter keyboard for an infinite amount of time will almost surely type a given text, such as the complete works of William Shakespeare. In fact, the monkey would almost surely type every possible finite text an infinite number of times.
Happiness in intelligent people is the rarest thing I know.
Ernest Hemingway
They are basically taking the same spot that TA Edison used so successfully The same idea was also used by the University of Manchester with automated lab assistants dubbed Adam and Eve https://www.chemistryworld.com...
After the AI update, google results went from being usable, to total unbelievable dogshit over night.
Unfortunately the user base seemed to go from tech savvy, to brain dead NPCs also over night, so nobody noticed.
Now I'm stuck here surrounded by shit robots. The NPCs won't believe that neural networks are still stupid toys like 20 years ago.
They don't understand how the media works, the hype cycle, only believe literally what they're told.
Corporate money will "fix" that "issue" and they will own results of automated processes.
Big news everyone!!! Millenial bro just invented the next big thing (because we're AWESOME)
seriously, how could any generation before us live without the internet and instagram? Are they dumb or what?
Liberals scoff at religion, and instead choose to worship Science and her chosen messengers known as Scientist. Now that we have artiricial intelligence doing science we can worship AI. Just as nonscientist must defer all thinking to to scientist. Now scientist can defer to artificial intelligence.
All must bow down to the alogorithm. Human agency will be eliminated in the brave
Creativity is over rated
Filters "useless" compounds. Um, ok. But what if the startling breakthrough takes place in one of those assumed useless compounds? No insight possible. No flash of brilliance. No serendipity. No human advantage possible. Remember junk DNA? It was until it wasn't.
E Proelio Veritas.
Feel free to start your own robot lab and give away the results for free.
Robots don't dream -- there are no brainwaves. Either it is trying materials at random or enumerating the permutations.
Who the fuck writes this garbage?
function DreamUp()
{
return Math.floor(Math.random() * NUM_MATERIALS);
}
Do you misuse phrases ("dream up") on purpose or do you really, really believe that humans dreaming up a system to streamline a research process actually means the new system then becomes conscious?
Note: My usage is correct, so my money's on you're doing it purposefully.
Won't somebody think of the poor grad students?!
The company believes it may find new compounds that could, among other things, "absorb pollution" (murder humans), "combat drug-resistant fungal infections" (murder humans), and "serve as more efficient optoelectronic components" (murder humans).
I applaud whoever took the founders mugshots, makes them all look like they themselves were forced to give this machine a body and just say they did it intentionally.