Data Scientist Romeo Kienzler shows you how to create a DeepLearning-based anomaly detector using DeepLearning4J and Apache SystemML on Apache Spark.

In this video:

Deep learning frameworks are popping up with some frequency, but only a few of them are suitable to run on clusters and use GPUs and supporting topologies (other than Feed-Forward) at the same time. DeepLearning4J, Apache SystemML, and TensorSpark do have these features and they don’t force you to learn new exotic programming languages. In addition, they also scale out on well established infrastructures like ApacheSpark.

In this live event replay, Data Scientist Romeo Kienzler introduces DeepLearning4J and Apache SystemML running atop Apache Spark, then he produces an example of the power of this combination – you’ll see how to create an anomaly detector for IoT sensor data with an LSTM auto-encoder neural network. Romeo also details how Apache SystemML uses cost-based optimisers for neural network training and how TensorSpark parallelizes TensorFlow on ApacheSpark.

Romeo Kienzler is a Senior Data Scientist and DeepLearning and AI Engineer for IBM Watson IoT and an IBM Certified Senior Architect who spends much of his waking life helping global clients solve their data analysis challenges. Romeo holds an MSc (ETH) in Computer Science with specialization in information systems, bioinformatics, and applied statistics from the Swiss Federal Institute of Technology. He is an Associate Professor of artificial intelligence and his current research focus is on cloud-scale machine learning and deep learning using open source technologies including R, Apache Spark, Apache SystemML, Apache Flink, DeepLearning4J, and TensorFlow.

Resources for you

Discovering Data Science with Romeo Kienzler
Tackle the most difficult challenges in data science.

Subscribe by email | Subscribe on YouTube

developerWorks Live Webcasts
Live coding demos, webinars, and “ask me anythings”.

Subscribe by email | Subscribe on YouTube

Apache Spark
Create algorithms to harness insight from complex data.

Join The Discussion

Your email address will not be published. Required fields are marked *