IISc scientists develop brain-inspired analog computing platform capable of storing, processing data Premium

IISc researchers develop brain-inspired analog computing platform with 16,500 conductance states, revolutionizing AI hardware and technology innovation.

Researchers at the Indian Institute of Science (IISc) have developed a brain-inspired analog computing platform capable of storing and processing data in an astonishing 16,500 conductance states within a molecular film. Published on September 11 in the journal Nature, this breakthrough represents a step forward over traditional digital computers in which data storage and processing are limited to just two states, the IISc said.

In a press release, the institute said that such a platform could potentially bring complex AI tasks, like training Large Language Models (LLMs), to personal devices like laptops and smartphones, thus taking us closer to democratising the development of AI tools. These developments are currently restricted to resource-heavy data centres, due to a lack of energy-efficient hardware. With silicon electronics nearing saturation, designing brain-inspired accelerators that can work alongside silicon chips to deliver faster, more efficient AI is also becoming crucial.

“Neuromorphic computing has had its fair share of unsolved challenges for over a decade,” explained Sreetosh Goswami, Assistant Professor at the Centre for Nano Science and Engineering (CeNSE), IISc, who led the research team. “With this discovery, we have almost nailed the perfect system – a rare feat," he said.

According to IISc, the fundamental operation underlying most AI algorithms is quite basic – matrix multiplication, a concept taught in high school maths. But in digital computers, these calculations hog a lot of energy. The platform developed by the IISc team drastically cuts down both the time and energy involved, making these calculations a lot faster and easier.

The molecular system at the heart of the platform was designed by Sreebrata Goswami, Visiting Professor at CeNSE. As molecules and ions wiggle and move within a material film, they create countless unique memory states, many of which have been inaccessible so far. Most digital devices are only able to access two states (high and low conductance), without being able to tap into the infinite number of intermediate states possible.

By using precisely timed voltage pulses, the IISc team found a way to effectively trace a much larger number of molecular movements, and map each of these to a distinct electrical signal, forming an extensive “molecular diary” of different states. “This project brought together the precision of electrical engineering with the creativity of chemistry, letting us control molecular kinetics very precisely inside an electronic circuit powered by nanosecond voltage pulses,” explained Goswami.

"Tapping into these tiny molecular changes allowed the team to create a highly precise and efficient neuromorphic accelerator, which can store and process data within the same location, similar to the human brain. Such accelerators can be seamlessly integrated with silicon circuits to boost their performance and energy efficiency," the IISc said.