SmartTensors AI: Unsupervised software big datasets
Boian Alexandrov and team used the SmartTensors AI Platform technology to extract latent features from a 3D simulation of an asteroid’s deep- water impact.
The SmartTensors AI Platform, developed at Los Alamos National Laboratory, is a scalable, unsupervised machine-learning software suite capable of identifying, extracting essential hidden features, and efficiently compressing information in massive datasets. SmartTensors autonomously analyzes and discovers hidden features, signatures and patterns otherwise undetectable and buried in tens of terabytes of data.
Big data surrounds us and is growing exponentially bigger, requiring massive storage and processing power. Crucial bits of information and signals are frequently overlooked and lost in the noise because traditional tools cannot process extra-large data and extract hidden signals from it. Researchers at Los Alamos have developed multilingual software, the SmartTensors AI Platform, to keep up with this onslaught of knowledge. The unique platform’s capabilities include latent feature extraction, dimension reduction, pattern recognition, anomaly detection, data compression, and text mining.
SmartTensors has been applied effectively and impactfully to medicine, disease spread analysis/prediction, energy extraction, carbon sequestration, climate change, economic analysis, infrastructure stability and national security. SmartTensors can be used on laptops, desktops, supercomputers and cloud platforms.