Biography

Peter Varhol is a well-known writer and speaker on software and technology topics, having authored dozens of articles and spoken at a number of industry conferences and webcasts. He has advanced degrees in computer science, applied mathematics, and psychology. Currently he has his own consulting company, Technology Strategy Research, specializing in AI, Big Data, and analytics, and is also a blogger and blog editor for Toptal, LLC. His past roles include technology journalist, software product manager, software developer, and university professor.

Description

While we refer to deep learning, genetic algorithms, and other modern AI techniques as intelligent, they really only represent advanced statistical techniques that simply correlate one set of independent variables to one or more dependent variables. That which we call intelligence is often only statistical correlation. If we want to get beyond simple correlation, we have to connect cause with effect. Determining causal relationships requires tried-and true scientific methods, that is, empirical and measurable evidence subject to testable explanations and predictions. To successfully test intelligent systems, we need to formally incorporate the language of hypotheses with objective evidence either supporting or rejecting those hypotheses. This presentation incorporates work by MIT Media Lab Professor Alex Pentland, Bayes Theory researcher Judea Pearl, and others to determine the difference between correlation and cause and effect, and how we can incorporate testing of AI applications to go beyond simple correctness of the algorithms to determining whether cause and effect exists.