Toggle light / dark theme

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 today in the journal Nature, this breakthrough represents a huge step forward over traditional digital computers in which data storage and processing are limited to just two states.

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 democratizing the development of AI tools. These developments are currently restricted to resource-heavy data centers, 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,” explains 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.”

Thanks to a serendipitous discovery and a lot of painstaking work, scientists can now build biohybrid molecules that combine the homing powers of DNA with the broad functional repertoire of proteins—without having to synthesize them one by one, researchers report in a new study. Using a naturally occurring process, laboratories can harness the existing molecule-building capacities of bacteria to generate vast libraries of potentially therapeutic DNA-protein hybrid molecules.

In a recent paper published in PRX Quantum, a team of researchers from Osaka University and RIKEN presented an approach to improve the fault-tolerance of color codes, a type of quantum error correction (QEC) code. Their method, known as Flagged Weight Optimization (FWO), targets the underlying challenges of color-code architectures, which historically suffer from lower thresholds under circuit-level noise. By optimizing the decoder weights based on the outcomes of flag qubits, this method improves the threshold values of color codes.

Color codes are an alternative to surface codes in quantum error correction that implement all Clifford gates transversally, making them a potential solution for low-overhead quantum computing, as noted by the paper. However, their practical use has been limited thus far by the relatively low fault-tolerance thresholds under circuit-level noise. Traditional methods of stabilizer measurement, which involve high-weight stabilizers acting on numerous qubits, introduce substantial circuit depth and errors, ultimately leading to lower overall performance.

The research team focused on two color-code lattices—the (4.8.8) and (6.6.6) color codes. The team noted that while these codes are considered topologically advantageous for QEC, their previous thresholds were relatively low, making them less effective for real-world applications. For example, the threshold for the (4.8.8) color code was previously around 0.14%, limiting its use in fault-tolerant computing.