Programming Frameworks for Deep Learning2018HT
Introductory lecture (half day), miniproject, demo/presentation session.
As this course focuses on programming and acceleration aspects, we will only shortly summarize the theory of neural networks and, where applicable, recommend to combine this course with another course on the theory of neural networks, such as Neural Networks and Deep Learning.
Graduate (CUGS, CIS, ...) students interested in the use and the structure of domain-specific programming frameworks for artificial neural networks and deep learning.
The course was last given
This is a new course.
The course studies the programming model, structure, optimization and
acceleration opportunities for some popular domain-specific programming
frameworks for deep learning, such as TensorFlow.
Programming in C++ and/or Python.
Some background in parallel and GPU/accelerator computing, such as TDDD56 or TDDC78.
Data structures and algorithms.
Some introductory course on machine learning including neural networks.
* Short recapitulation of fundamental issues in machine learning, feed-forward
neural networks, backpropagation learning algorithm.
* Study of some popular domain-specific programming frameworks for neural networks and deep learning, such as TensorFlow, Caffe, Dlib.
- Programming constructs, Framework structure, Examples. Classification.
* Opportunities and techniques for accelerated learning and execution on parallel systems (multicore, GPGPU).
* Individual miniproject by participants.
Presentation of miniproject results.
To be announced.
* Active participation in the introductory lecture and presentation sessions
* Successful individual miniproject and presentation
3hp if both examination moments are fulfilled.
This course could complement the announced (HT2018) course Neural Networks and Deep Learning, but is not dependent on it.
Page responsible: Director of Graduate Studies