Time: 14:00 on January 10, 2017
Place: room 816, Ta Quang Buu Library, HUST
Presenter: Truyen Tran, Deakin University, Australia
Title: Deep architecture engineering
Abstract:
In 2016, deep learning took over the Internet with massive investment and intensive hunting for top talents by all major Internet players. The main power of deep learning is the ability to learn from raw signals without tedious feature engineering. At the heart of current deep learning in practice is architecture engineering — the science and art of designing neural nets to meet the problem structure at hand. In this talk, I will cover our recent effort at PRaDA (Deakin University) in designing deep neural nets for gates, events, sequences, permutations, matrices, graphs and relations. The applied domains include healthcare, software engineering, social media and vision.
More about speaker: Truyen Tran is a lecturer at Deakin University, Australia. His current research interests are in deep learning, healthcare, bioinformatics and software analytics. He received multiple paper awards and prizes including UAI 2009, CRESP 2014, Kaggle 2014, PAKDD 2015, ACM SIGSOFT 2015 and ADMD 2016. He obtained a Bachelor of Science from University of Melbourne and a PhD in Computer Science from Curtin University in 2001 and 2008, respectively.