Current Advances in Deep and Recurrent Neural Networks
Dozent | Dr. Sebastian Otte |
Zeit | Fr 10 c. t. bis 12 Uhr Seminar |
Umfang | 3 LP |
Beginn | 20.04.2017 |
Ort | Informatik, Sand 14, Poolraum, C 412 |
Prüfungsfach | |
Modul |
Topics
- Very Deep Networks
- Densely Connected Convolutional Networks
- Generative Adversarial Networks I
- Generative Adversarial Networks II: Wasserstein GAN
- Generative Adversarial Networks II: Synthesizing Audio
- Generative Adversarial Networks II: Image-to-Image Translation
- Single Shot MultiBox Detector
- Capsule Networks
- Recurrent Weighted Average Networks
- Recurrent Highway Networks
- Nested LSTMs
- Optimization and Regularization: SGD vs. Adam and Co.
- Optimization and Regularization: Recurrent Batch Normalization
- Optimization and Regularization: Batch Kalman Normalization
- Spherical CNNs
- Variational Autoencoders I
- Variational Autoencoders II: Recurrent VAs for music
- Wasserstein Auto-Encoders
- Bayesian Deep Learning
- Uncertainty-Aware Deep Learning
- Efficient Deep Learning: ShuffleNet
- Efficient Deep Learning: Multi-Scale Dense Networks
- Efficient Deep Learning: Integers in Deep Networks
- Fooling Deep Networks
- Understanding Deep Models
- Ethics in Artificial Intelligence and Deep Learning