How to Determine the Optimal Anomaly Detection Method For Your Application

Sunday, May 19, 2019, 13:00 - 16:30, Royal Palm

Cynthia Freeman, Ian Beaver

Verint Intelligent Self-Service
Abstract: An anomaly in a time series is a pattern that does not conform to past patterns of behavior in the series. Anomalies are important to detect as they can indicate events such as pending sensor failures, unexpected environmental conditions, or malicious activity. Unfortunately, there is no one best way to detect all anomalies across a variety of domains; such a methodology is a myth given that time series can display a wide range of behaviors. In addition, what behavior is anomalous can differ from application to application. In this tutorial, we introduce a framework that helps you determine the best anomaly detection method for your application based on the characteristics the time series possesses. For example, some anomaly detection methods will never adapt after a concept drift, predicting every point afterwards to be an anomaly. Some anomaly detection methods require interpolation of missing time steps beforehand while other can handle missing or nonuniform time steps innately. Participants will get hands-on experience applying various anomaly detection methods to several datasets exhibiting different kinds of behaviors. We will then discuss how best to evaluate them (precision, recall, F-score, NAB score, etc.) and choose an appropriate method specific to the time series’ behaviors.

Neural Nets 101

Sunday, May 19, 2019, 13:00 - 16:30, Cypress

David Bisant

Central Security Service
Abstract: Neural networks have become popular in the last few years due to the success that deep learning methods have had on a number of high profile image and speech problems. It is important to understand the basics in order to recognize what is “hype” and what is reality. Neural networks are powerful and flexible methods but it is also important to know their limitations.This tutorial will cover an introduction to modern neural networks and how they are applied to problems in artificial intelligence. Basic terminology, history, application methods, and application case studies will be covered. Modern topics such as deep learning will be covered. The material will be filtered and summarized for the novice. Further reading and software packages and frameworks will also be discussed.
Part 1:- Biological neurons- Artificial neurons and network structures- A history of artificial neural network (ANN) development- Machine Learning and problems where ANN’s can be applied
Part 2:- Learning algorithms for neural networks- Application case studies- Deep learning and deep belief networks- A brief mention of neural network hardware- Neural networks compared to other machine learning methods- Further reading