Researchers have developed an intelligent system for 'tuning' powered prosthetic knees, allowing patients to walk comfortably with the prosthetic device in minutes, rather than the hours necessary if the device is tuned by a trained clinical practitioner. The system is the first to rely solely on reinforcement learning to tune the robotic prosthesis.
A recent study has found that a specific gene in cancerous prostate tumors indicates when patients are at high-risk for the cancer to spread, suggesting that targeting this gene can help patients live longer.
A team of scientists has determined the number of asteroid impacts on the moon and Earth increased by two to three times starting around 290 million years ago. Previous theories held that there were fewer craters on both objects dating back to before that time because they had disappeared due to erosion. The new findings claim that there were simply fewer asteroid impacts during that earlier period.
Scientists have traditionally thought that DNA binding proteins use patterns in the genome's code of As, Cs, Ts, and Gs to guide them to the right location, with a given protein only binding to a specific sequence of letters. In a new study, published in Cell Systems, scientists at the Gladstone Institutes discovered that proteins must rely on another clue to know where to bind: the DNA's three-dimensional shape.
By combining deep learning algorithms and statistical methods, investigators from the Institute of Evolutionary Biology (IBE), the Centro Nacional de Análisis Genómico (CNAG-CRG) of the Centre for Genomic Regulation (CRG) and the Genomics Institute at the University of Tartu have identified, in the genome of Asiatic individuals, the footprint of a new hominid who cross bred with its ancestors tens of thousands of years ago.
MIT researchers are hoping to advance the democratization of data science with a new tool for nonstatisticians that automatically generates models for analyzing raw data.
In what the authors believe is the first documented comparison of several real-time infectious disease forecasting models by different teams across many seasons, five research groups report this week that a majority of models consistently showed higher accuracy than historical baseline models.
Numenta researchers propose a broad framework for understanding what the neocortex does and how it works. 'The Thousand Brains Theory of Intelligence' proposes that every part of the neocortex learns complete models of objects and concepts, rather than learning one model of the world.
Idtracker.ai is a mix of conventional algorithms and artificial intelligence developed at the Champalimaud Centre for the Unknown. From the video footage of a moving crowd composed of dozens of individuals, it learns to identify each and every individual in that crowd.
One of the greatest challenges in saving endangered species is to predict if an animal population will die out. Accurate and reliable models are crucial for conservationists.