A couple of days ago DailyTech brought probabilistic computing under the spotlight. The reporter talks about the work of Krishna Palem at Rice University, he and his team of researchers are testing a new probabilistic chip design that offers seven times the computing speed, at 1/30th of the power consumption of a traditional CPU. However, the drawback is that these chips are less accurate, this makes probabilistic chips undesirable for applications where high accuracy is needed but for tasks like graphics processing where a tiny bit of noise doesn't matter they are perfect. Furthermore, probabilistic processors also hold great promise to boost the energy efficiency of mobile devices like handsets and notebooks.
His probabilistic CPU chip ran seven times as fast as traditional circuits and consumed only 1/30th of the energy. The results match or even exceed those predict by his mathematical models and computer simulations. He states, "The results were far greater than we expected. At first, I almost couldn’t believe them. I spent several sleepless nights verifying the results."
Professor Palem's chips could revolutionize fields such as computer-generated graphics, content streaming, and other applications that would accept a tradeoff for precision for increases in computing speed and decreases in power. Many experts in these fields which had been previously reticent about the idea of probabilistic computing have been convinced by Professor Palem's results.
Al Barr, a computer scientist at the California Institute of Technology, is among the new believers. He acknowledges former doubts, stating, "Initially there was definitely a lot of skepticism."