- This event has passed.
Using Deep Learning to do Continuous Scoring in Practical Applications
January 25, 2016 @ 6:30 pm
Greg Makowski, Director of Data Science, LigaDATA
*** Bring ID (e.g. Driver’s License) for eBay Security ***
6:30 Doors Open, Food & Networking
*** Please arrive by 7 PM due to Security ***
Please RSVP or un-RSVP so we can more accurately plan for our food for the number of people.
Greg Makowski will talk about using Deep Learning to do real-time scoring in practical applications. He will talk about the current state of the art and what will be done in the future. The talk is based on both his experience in the front lines doing continuous and real-time analysis for banks and other enterprises.
The talk will cover a brief review of neural network basics and the following types of neural network deep learning:
* autocorrelational – unsupervised learning for extracting features. He will describe how additional layers build complexity in the feature extraction.
* convolutional – how to detect shift invariant patterns in various data sources. Horizontal shift invariant detection applies to signals like speech recognition or IoT data. Horizontal and vertical shift invariance applies to images or videos, for faces or self driving cars
* discuss details of applying deep net systems for continuous or real time scoring
* reinforcement learning or Q Learning – such as learning how to play Atari video games
* continuous space word models – such as word2vec, skipgram training, NLP understanding and translation
Greg Makowski is the Director of Data Science at LigaDATA, a Series A funded startup offering both big data and data mining consulting, as well as supporting big data open source systems and <a>www.Kamanja.org</a> Greg has deployed about 90 data mining models since 1992. Greg has prototyped and deployed four new enterprise applications with embedded data mining, and worked in a variety of verticals including financial services, fraud detection, web behavior, retail supply chain and targeted marketing. Greg first deployed a Convolutional Neural Network in 1991, for his thesis “Time Delay Neural Networks for Speech Recognition.” following the work of LeCun and others.
Event page provided by ACM