Skip to content

Changes

Page history
Update JULAIN Journal Club authored by susanne wenzel's avatar susanne wenzel
Hide whitespace changes
Inline Side-by-side
JULAIN-Journal-Club.md
View page @ ee98977a
......@@ -14,11 +14,15 @@ If you’re interested in more details about the Journal Club, please subscribe
## Next Meeting
### Monday 18 May 10-11:30am Variational Autoencoders
Venue: if possible f2f JSC meetingroom 2, building 16.3; room 315 <br>
alternatively online: https://webconf.fz-juelich.de/b/wen-mym-pj7
* Original VAE paper<br>
Auto-Encoding Variational Bayes<br>
Diederik P Kingma, Max Welling, ICLR 2014<br>
https://arxiv.org/abs/1312.6114 <br>
https://openreview.net/forum?id=33X9fd2-9FyZd <br>
https://openreview.net/forum?id=33X9fd2-9FyZd
* Recent VAE review / tutorial<br>
An Introduction to Variational Autoencoders<br>
Diederik P. Kingma, Max Welling (2019) Foundations and Trends in Machine Learning. 12. 307-392. 10.1561/2200000056. <br>
......@@ -26,6 +30,52 @@ https://arxiv.org/abs/1906.02691<br>
## Schedule for upcoming Meetings
### 15 June 2020 - Generative Adversarial Networks (and VAE)
Venue: if possible f2f JSC meetingroom 2, building 16.3; room 315 <br>
alternatively online: https://webconf.fz-juelich.de/b/wen-mym-pj7
* Original GAN paper:<br>
Generative Adversarial Networks<br>
Goodfellow, Bengio et al, NIPS 2014<br>
https://arxiv.org/abs/1406.2661 <br>
https://papers.nips.cc/paper/5423-generative-adversarial-nets.html
* On relation GAN, VAE:<br>
On Unifying Deep Generative Models<br>
Zhiting Hu, Zichao Yang, Ruslan Salakhutdinov, Eric P. Xing, ICLR 2017<br>
https://arxiv.org/abs/1706.00550 <br>
https://openreview.net/forum?id=rylSzl-R-
### 20 July 2020 - Summer break
### 17 August 2020 - Attention Networks
Venue: if possible f2f JSC meetingroom 2, building 16.3; room 315 <br>
alternatively online: https://webconf.fz-juelich.de/b/wen-mym-pj7
* Attention Is All You Need<br>
Vaswani, Shazeer, Parmar, Uszkoreit, Jones, Gomez, Kaiser, Polosukhin, NIPS 2017 <br>
https://papers.nips.cc/paper/7181-attention-is-all-you-need.pdf
* Self-Attention Generative Adversarial Networks<br>
Zhang, Goodfellow, Metaxas, Odena, ICML 2019<br>
https://arxiv.org/abs/1805.08318
### 21 September 2020 Self-Supervised Visual Representation Learning
Venue: if possible f2f JSC meetingroom 2, building 16.3; room 315 <br>
alternatively online: https://webconf.fz-juelich.de/b/wen-mym-pj7
* A Simple Framework for Contrastive Learning of Visual Representations <br>
T. Chen, S. Kornblith, M. Norouzi, and G. Hinton, 2020<br>
http://arxiv.org/abs/2002.05709
* Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey <br>
L. Jing and Y. Tian, CVPR2019<br>
http://arxiv.org/abs/1902.06162
### Monday 21 September 10-11:30am - Self-Supervised Visual Representation Learning
Venue: tba
......@@ -40,10 +90,8 @@ The first paper presents a state-of-the-art approach for self-supervised learnin
## Past Meetings
### Monday 20 April 10-11:30am - Learning discrete representations from data
### Monday 20 April - Learning discrete representations from data
Venue: JSC meetingroom 2, building 16.3, room 315
* A. van den Oord, O. Vinyals, K. Kavukcuoglu, Neural Discrete Representation Learning, NeurIPS 2017<br>
https://arxiv.org/abs/1711.00937
* A. Razavi, A. van den Oord, O. Vinyals, Generating Diverse High-Fidelity Images with VQ-VAE-2, NeurIPS 2019<br>
......@@ -52,15 +100,10 @@ https://arxiv.org/abs/1906.00446
Those two papers are about learning discrete representations from data by taking inspiration from vector quantization. Learning discrete representations using neural networks is challenging and can helpful for tasks such as compression, planning, reasoning, and can be potentially more interpretable than continuous ones. In the two papers use those learned discrete representations to build autoregressive generative models on image, sound, and video. The second paper (Generating Diverse High-Fidelity Images with VQ-VAE-2) is basically a sequel of the first (Neural Discrete Representation Learning) where they scale the models to bigger datasets and images (up to 1024x1024 resolution).
### Monday 30 March 10-11:30am - Full-Resolution Residual Networks for Semantic Segmentation
**Replacement date for canceled meeting at Monday 16 March 10-11:30am**
### Monday 30 March - Full-Resolution Residual Networks for Semantic Segmentation
virtual meeting via dfnconf: https://conf.dfn.de/webapp/conference/97977564<br>
alternative link, if dfn is down: https://us04web.zoom.us/j/433015211
**Replacement date for canceled meeting at Monday 16 March**
Venue: **INM-1 Seminar room**, building 15.9, room 4001b
* Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes (CVPR'17 Oral)<br>
T. Pohlen, A. Hermans, M. Mathias, and B. Leibe<br>
Paper: http://arxiv.org/abs/1611.08323 <br>
......@@ -72,9 +115,7 @@ Code: https://paperswithcode.com/paper/deep-high-resolution-representation-learn
Most common architectures for semantic segmentation consisting of an encoder and decoder part (e.g. U-Net) heavily reduce the spatial dimension of input images and may loose important details or fail to localize precisely. The proposed papers present full-resolution networks, which try to preserve high-resolution features throughout the network and improve localization accuracy.
### Monday 17 February 10-11:30am - Speech Recognition
Venue: **INM-1 Seminar room**, building 15.9, room 4001b
### Monday 17 February - Speech Recognition
* Deep Speech 2: End-to-End Speech Recognition in English and Mandarin, Amodei et al., 2015<br>
https://arxiv.org/abs/1512.02595<br>
......@@ -84,20 +125,16 @@ Speech Recognition with Deep Recurrent Neural Networks<br>
Alex Graves, Abdel-rahman Mohamed, Geoffrey Hinton, 2013<br>
https://arxiv.org/abs/1303.5778
### Monday 20 January 10-11:30am
### Monday 20 January
Venue: **INM-1 Seminar room**, building 15.9, room 4001b
* Multi-Context Recurrent Neural Networks for Time Series Applications <br>
https://publications.waset.org/3524/pdf
* Global Sparse Momentum SGD for Pruning Very Deep Neural Networks <br>
https://arxiv.org/pdf/1909.12778v3.pdf
### Monday 16 December 10-11am
### Monday 16 December
Venue: **JSC, Rotunda**, building 16.4, room 301
Our first Journal Club will cover two papers from ICCV 2019 about GANs.
* SinGAN: Learning a Generative Model from a Single Natural Image (Best Paper Award) <br> http://openaccess.thecvf.com/content_ICCV_2019/papers/Shaham_SinGAN_Learning_a_Generative_Model_From_a_Single_Natural_Image_ICCV_2019_paper.pdf
* COCO-GAN: Generation by Parts via Conditional Coordinating <br>
......