GTC-DC 2019: GPU-Accelerated IIoT Data Science and Machine Learning in an Enterprise Data Lake (Presented by Lockheed Martin)
Antonio DeMarco, Lockheed Martin; Bryan Massie, Lockheed Martin
We’ll discuss how Lockheed Martin’s critical assets in factories produce a wealth of valuable data, not just aircrafts. Using NVIDIA GPUs, we’re extracting value from this telemetry data through neural network models. We’ll give an overview of our efforts and the technology we’re leveraging to harvest and store the data. We’ll also examine a long short-term memory recurrent neural network model that we’ve implemented for automated anomaly detection in the time series data.