University of Kentucky

MA 721: Topics in Numerical Analysis: Deep Learning

Fall 2017

MWF 1:00 pm - 1:50 pm, CB 347

List of Reading for Class Presentations

Please prepare slides for your presentation and submit it by email at least three days before the planned presentation date for comments/grading.

Paper Presenter Week
Introduction to Tensorflow Liu Liu 10/6
Introduction to Theano ? ?
Maxout Networks Jiho Noh/td> 10/9
On the Number of Linear Regions of Deep Neural Networks ? 10/9
Improving the convergence of back-propagation learning with second order methods Kasey Bray 10/11
ADADELTA: An Adaptive Learning Rate Method Daniel Cotter 10/16
Adam: A Method for Stochastic Optimization Xinxin Zuo 10/16
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift Devin Willmott 10/16
Identifying and attacking the saddle point problem in high-dimensional non-convex optimization Ryan Peters 10/23
Understanding the difficulty of training deep feedforward neural networks ? 10/23
Exact solutions to the nonlinear dynamics of learning in deep linear neural networks ? 10/30
Random Walk Initialization for Training Very Deep Feedforward Networks ? 10/30
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification Gayan Maduranga TBA
All you need is a good init ? TBA
Visualizing and Understanding Convolutional Networks Gongbo Liang TBA
Going deeper with convolutions ? TBA
Network In Network ? TBA
Deep Residual Learning for Image Recognition Jinpeng Liu TBA
Long Short-term Memory Derek Jones TBA
Adaptive nonlinear system identification with echo state networks. ? TBA
Orthogonal Recurrent Neural Networks with Scaled Cayley Transform Kyle Helfrich TBA