Welcome to PEARC20! PEARC20’s theme is “Catch the Wave.” This year’s theme embodies the spirit of the community’s drive to stay on pace and in front of all the new waves in technology, analytics, and a globally connected and diverse workforce. We look forward to this year’s PEARC20 virtual meeting, where we can share scientific discovery and craft the future infrastructure.
The conference will be held in Pacific Time (PT) and the times listed below are in Pacific Time.
The connection information for all PEARC20 workshops, tutorials, plenaries, track presentations, BOFs, Posters, Visualization Showcase, and other affiliated events, are in the PEARC20 virtual conference platform, Brella. If you have issues joining Brella, please email pearcinfo@googlegroups.com.
There are many capable Deep Neural Networ software frameworks available freely. Most of them typically provide a python user interface which lets a user quickly train and test his/her dataset without bothering how DNN actually works. Using these DNN frameworks is clearly advantageous to a majority of users, allowing them spending more time on their own projects without worrying the integrity of their solution because these DNN frameworks are well known and accepted as standard in the data science community. However, in some cases, when a researcher may in fact want to find out why and how a method of one of these open source DNN frameworks works or fails, particularly with respect to his/her problems, it will be a difficult task for the researcher because he/she lacks the experience to understand how in general the DNN algorithm is implemented in the backend compute engine which is in general written in C++. As such this tutorial intends to prepare the users of such needs, broadening their knowledge in DNN via a step by step illustration starting from the mathematical formulation, numerical algebra interpretation, to programming implementation.
