SS 2019

Mathematics of Deep Learning

 

Schedule


Date Topic Title Presenters Links Exercises Notes
15 Apr Introduction Introduction Antoine and Jonas

slides on formalities

29 Apr Foundations Derivations and gradients Antoine and Jonas

slides on derivations and gradients

06 May More foundations Gradient descent and backpropagation Antoine and Jonas

exercise
exercise with solution headlines.
neural network in a spreadsheet
POS tagging results

13 May LSTMs Liu et al. (2018): LSTMs Exploit Linguistic Attributes of Data Sasha, Johannes

presentation's slides

20 May LSTMs Levy et al. (2018): Long Short-Term Memory as a Dynamically Computed Element-wise Weighted Sum Saumya, Sara

prensentation's slides

27 May Convolutional neural nets Cireșan et al. (2012): Multi Column Deep Neural Network for Image Classification Martin K

03 Jun Attention Bahdanau et al. (2014): Neural Machine Translation by Jointly Learning to Align and Translate Christian S, Anna

10 Jun Holiday Pentacost

17 Jun Cancelled

24 Jun Tree-structured networks Tai et al. (2015): Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks Udeh

01 Jul Tree-structured networks Vaswani et al. (2018): Attention Is All You Need Anil

presentation's slides (ppt)

08 Jul Tree-structured networks Dyer et al. (2016): Recurrent neural network grammars Christian B

08 Jul Tree-structured networks Kuncoro et al. (2018): LSTMs Can Learn Syntax-Sensitive Dependencies Well, But Modeling Structure Makes Them Better Atakan

11 Jul Implementation Baydin et al. (2018): Automatic Differentiation in Machine Learning: a Survey Marius