Deep Learning has become a major breakthrough in Artificial Intelligence in recent years. This talk will include a study on implementing L-BFGS, which is a popular optimization algorithm for parameter estimation in machine learning. The speaker will explain how HPCC systems provides the platform as well as the programming language (ECL) in order to implement a parallelized and distributed version of L-BFGS algorithm which work on Big Data. L-BFGS algorithm can be used in not only implementing deep learning algorithms, but also any other Machine Learning algorithm which needs estimating parameters based on optimizing a cost function, such as SVM, Logistic regression, etc.,. At the end the implementations of Sparse Autoencoder algorithm and Softmax classifier are explained as a demonstration on how to use L-BFGS for parameter estimation in such algorithms.
Maryam M. Najafabadi received her B.S. degree in Computer Science from Isfahan University of Technology and her M.S. degree in Artificial Intelligence from Amirkabir University of Technology in 2008 and 2011 respectively. Currently, she is a Ph.D. candidate in the Department of Computer and Electrical Engineering and Computer Science at Florida Atlantic University. Her research interests include Data Mining and Machine Learning. Since 2013 Maryam has been involved in the collaborative research between FAU and LexisNexis where her primary task is implementing machine learning algorithms, focusing on Neural Networks and Deep Learning.