Amazing stuff! Potentially a breakthrough in analog deep learning!
"Engineers at MIT have developed a new type of artificial synapse that’s extremely energy efficient and ultra-fast, processing data a million times faster than synapses in the human brain. ..."
"... A new area of artificial intelligence called analog deep learning promises faster computation with a fraction of the energy usage. Programmable resistors are the key building blocks in analog deep learning, just like transistors are the core elements for digital processors. By repeating arrays of programmable resistors in complex layers, researchers can create a network of analog artificial “neurons” and “synapses” that execute computations just like a digital neural network. ...
Analog deep learning is faster and more energy-efficient than its digital counterpart for two main reasons. “First, computation is performed in memory, so enormous loads of data are not transferred back and forth from memory to a processor.” Analog processors also conduct operations in parallel. ...
Analog deep learning is faster and more energy-efficient than its digital counterpart for two main reasons. “First, computation is performed in memory, so enormous loads of data are not transferred back and forth from memory to a processor.” Analog processors also conduct operations in parallel. ...
To develop a super-fast and highly energy efficient programmable protonic resistor, the researchers looked to different materials for the electrolyte. ... inorganic phosphosilicate glass (PSG).
PSG is basically silicon dioxide, which is the powdery desiccant material found in tiny bags that come in the box with new furniture to remove moisture. It is studied as a proton conductor under humidified conditions for fuel cells. It is also the most well-known oxide used in silicon processing. To make PSG, a tiny bit of phosphorus is added to the silicon to give it special characteristics for proton conduction. ..."
"The speed of biological information processing in neurons and synapses is limited by the aqueous medium through which weak action potentials of about 100 millivolts propagate over milliseconds. Above 1.23 volts, liquid water decomposes. Artificial solid-state neurons are not limited by such time and voltage constraints and can also be fabricated at the nanoscale, 1000 times smaller than their biological counterparts. Using complementary metal-oxide semiconductor–compatible materials, Onen et al. prototyped nanoscale protonic programmable resistors that can withstand high electric fields of around 10 megavolts per centimeter and which have energy-efficient modulation characteristics at room temperature. The proposed devices are 10,000 times faster than biological synapses and offer a promising direction for implementing various applications that can benefit from fast ionic motion"
From the abstract:
"Nanoscale ionic programmable resistors for analog deep learning are 1000 times smaller than biological cells, but it is not yet clear how much faster they can be relative to neurons and synapses. Scaling analyses of ionic transport and charge-transfer reaction rates point to operation in the nonlinear regime, where extreme electric fields are present within the solid electrolyte and its interfaces. In this work, we generated silicon-compatible nanoscale protonic programmable resistors with highly desirable characteristics under extreme electric fields. This operation regime enabled controlled shuttling and intercalation of protons in nanoseconds at room temperature in an energy-efficient manner. The devices showed symmetric, linear, and reversible modulation characteristics with many conductance states covering a 20× dynamic range. Thus, the space-time-energy performance of the all–solid-state artificial synapses can greatly exceed that of their biological counterparts."
New hardware offers faster computation for artificial intelligence, with much less energy Engineers working on “analog deep learning” have found a way to propel protons through solids at unprecedented speeds.
Nanosecond protonic programmable resistors for analog deep learning (no public access)
No comments:
Post a Comment