
Google Claims To Be Using AI To Design Computer Chips In Under 6 Hours
Google claims to have created a deep reinforcement learning software that can produce AI chips much faster than humans can, according to a paper in the Nature journal on Wednesday.
Essentially, Google is creating AI design chips that can draw up a “floorplan” for more advanced AI systems through an intricate process that involves carefully placing components, including CPUs, GPUs and memory cores, in relative positions across a silicon die. It typically requires months of intense effort by engineers to create these floorplans, but Google’s latest AI learning system, which is equipped with a learning program trained on a dataset of 10,000 varying floorplans, can produce manufacturable layouts in a much faster timeframe.
“In under six hours, our method automatically generates chip floorplans that are superior or comparable to those produced by humans in all key metrics, including power consumption, performance and chip area,” the paper reads. “To achieve this, we pose chip floorplanning as a reinforcement learning problem, and develop an edge-based graph convolutional neural network architecture capable of learning rich and transferable representations of the chip.”
The paper notes that this new development has “major implications” for the chip industry’s advancement, allowing companies to more swiftly test potential designs and create custom chips for specified tasks.
You may also like
TikTok Begins Testing ‘TikTok Stories’ That Disappear After 24 Hours
TikTok, the widely popular short-form video app, might soon be adding 24-hour “stories” to its i
Facebook Exploring New Prompt Against Extremist Content
Facebook is now testing new features aimed at helping users who have been related to extremist conte
Clubhouse Is Officially Open to the Public
More than a year after its initial launch, Clubhouse has officially moved out of its beta phase, dit
Post a comment