Machine learning studies need colossal power to process massive datasets and train neural networks to reach high accuracies, which have become gradually unsustainable. Limited by the von Neumann bottleneck, current computing architectures and methods fuel this high power consumption. Here, we present an analog computing method that harnesses chaotic nonlinear attractors to perform machine learning tasks with low power consumption. Inspired by neuromorphic computing, our model is a programmable, versatile, and generalized platform for machine learning tasks. Our mode provides exceptional performance in clustering by utilizing chaotic attractors' nonlinear mapping and sensitivity to initial conditions. When deployed as a simple analog device, it only requires milliwatt-scale power levels while being on par with current machine learning techniques. We demonstrate low errors and high accuracies with our model for regression and classification-based learning tasks.

Are you nervous about the pace of technological change? If you're running a profitable business, and would like to keep it that way, you should be worried about being out-innovated by a nimbler startup leveraging the latest tech. However, while machine learning and AI will play a major role in separating the winners from the losers, it won't be the only factor at play, according to a pair of MIT researchers. It's hard to overstate the advances being made in the world of data science, which is closely linked to AI and is ultimately the source of many of the digital transformations occurring right now. During his keynote address at the Dataworks Summit earlier this month, Hortonworks CEO Rob Bearden said data science will be "the transformational next leg of the journey" for everybody.