A Survey of Deep Learning Techniques in Data Management

PhD Qualifying Examination


Title: "A Survey of Deep Learning Techniques in Data Management"

by

Mr. Qiyu LIU


Abstract:

Over the past decade, machine learning, especially deep learning 
techniques, has achieved a breakthrough and opened a new paradigm for 
people to re-examine the power of computation. At the same time, the great 
advance in hardware and architecture enables a modern computer, even a 
personal desktop, to be equipped with powerful Single Instruction Multiple 
Data (SIMD) capabilities provided by CPU, GPU, and even Tensor Processing 
Unit (TPU). The advances made in computation power not only make 
large-scale machine learning models tractable but also inspire the 
database community to re-inspect the data management techniques used for 
decades including data indexing, query optimization and data structures 
like Hash table and Bloom filter.

In this survey, we conduct a literature review for this new but incredibly 
growing research direction, that is, introducing the deep learning power 
to solve data management challenges. We discuss the state-of-the-art deep 
learning techniques, the core challenges in the data management area, and 
their combination. Through this survey, we aim at providing a general view 
of the mainstream methodologies in "learning+DB" and introducing the 
possible future research possibilities.


Date:			Friday, 29 November 2019

Time:                  	9:00am - 11:00am

Venue:                  Room CYTG001 (CYT Building)
                         Lifts 35/36

Committee Members:	Prof. Lei Chen (Supervisor)
 			Prof. Ke Yi (Chairperson)
 			Dr. Yangqiu Song
 			Prof. Chiew-Lan Tai


**** ALL are Welcome ****