Nvidia DLI – Applications of AI for Predictive Maintenance
Dr. Laszlo Kovacs, kovacs.laszlo@inf.unideb.hu
You’ll learn how to identify anomalies and failures in time series data, estimate the remaining useful life of the corresponding parts, and use this information to map anomalies to failure conditions. You’ll be able to leverage predictive maintenance to manage failures and avoid costly unplanned downtimes. To begin, you’ll learn the key challenges around identifying anomalies that can lead to costly breakdowns. We’ll discuss how you can leverage your company’s time-series data to predict outcomes using machine learning classification models with XGBoost. Then, you’ll learn how to apply predictive maintenance procedures by using an LSTM-based model to predict the failure of a device and avoid downtime. Finally, you will experiment with autoencoders to detect anomalies by using the time series sequences from the previous steps. At the conclusion of the workshop, you’ll learn how to: • Predict part failures using machine learning classification models with XGBoost • Train GPU LSTM-based models using Keras and TensorFlow for failure prediction in time series • Detect anomalies using an autoencoder and Seq2Seq models • Experiment with generative adversarial network (GAN) models to detect anomalies
Learning Objectives
Introductory
8 hours
Short Course
Free of charge for university students and staff. An understanding of fundamental programming concepts in Python such as functions, loops, dictionaries, and arrays is a prerequisite.
Training XGBoost Models with RAPIDS for Time Series (120 mins), Training LSTM Models Using Keras and TensorFlow for Time Series (120), Training Autoencoders for Anomaly Detection (120 mins)
14.12.2024 9:00– 14.12.2024 17:00 CET
English
online
Upon successful completion of the assessment, the participant will receive an Nvidia Certificate of Competency.