New Hardware Delivers Faster Computation for Artificial Intelligence, With Much Less Energy MIT engineers working on “analog deep learning” have found a way to propel protons through solids at unprecedented speeds. The amount of time, effort, and money needed to train ever-more-complex neural ne
This illustration shows an analog deep learning processor powered by ultra-fast protonics. Credit: Ella Maru Studio, Murat Onenengineers working on “analog deep learning” have found a way to propel protons through solids at unprecedented speeds.
Furthermore, this inorganic material also makes the resistor incredibly energy-efficient. Unlike materials used in the earlier version of their device, the new material is compatible with silicon fabrication techniques. This change has enabled fabricating devices at the nanometer scale and could pave the way for integration into commercial computing hardware for deep-learning applications.
These programmable resistors drastically increase the speed at which a neural network is trained, while vastly reducing the cost and energy to perform that training. This could help researchers develop deep learning models much more quickly, which could then be applied in uses like self-driving cars, fraud detection, or medical image analysis.
In the human brain, learning happens due to the strengthening and weakening of connections between neurons, called synapses. Deep neural networks have long adopted this strategy, where the network weights are programmed through training algorithms. In the case of this new processor, increasing and decreasing the electrical conductance of protonic resistors enables analog machine learning.
Onen hypothesized that an optimized PSG could have a high proton conductivity at room temperature without the need for water, which would make it an ideal solid electrolyte for this application. He was right.PSG enables ultrafast proton movement because it contains a multitude of nanometer-sized pores whose surfaces provide paths for proton diffusion. It can also withstand very strong, pulsed electric fields.
Thanks to the insulating properties of PSG, almost no electric current passes through the material as protons move. This makes the device extremely energy efficient, Onen adds.
Norge Siste Nytt, Norge Overskrifter
Similar News:Du kan også lese nyheter som ligner på denne som vi har samlet inn fra andre nyhetskilder.
Rising Christian group We the Kingdom announce new album from New York's Times SquareWe the Kingdom announces its new studio album Friday following a performance at the 'Fox & Friends All-American Summer Concert Series.'
Les mer »
New Jersey Lottery Festival Of Ballooning Underway In Readington, New JerseyHAPPENING TODAY🎈🎈 It's the final day of the NJ Lottery Festival of Ballooning -- the largest balloon and music festival in North America
Les mer »
Mother says captured U.S. soldiers from Alabama moved to new prisonThe mother of retired Sgt. Alex Drueke, a former U.S. soldier captured in Ukraine, said he and Andy Huynh have been moved to a prison in an unspecified location.
Les mer »
Unlike Kelis, Kevin Aviance is grateful Beyoncé sampled him on new albumAviance said he didn’t get a heads up that the superstar was using part of his X-rated 1996 track on new song “Pure/Honey,” but was overwhelmed with joy.
Les mer »
2023 Porsche 911 GT3 R race car revealed, packs new 4.2-liter flat-6The race car from which Porsche will derive its next 911 GT3 RS has made its debut.
Les mer »
New Carrollton business park will bring six buildingsSan Francisco-based Prologis plans to build six warehouses in the Carrollton project.
Les mer »