Lecturer: Dietrich Klakow
Location: to be announced
Time: block course in the fall break 2019
Application for participation:
Please register here until May 1st.
Unordered List of Topics:
- Generating multiple objects at spatially distinct locations ( link to paper )
- h-detach: Modifying the LSTM Gradient Towards Better Optimization ( link to paper )
- Compositional attention networks for machine reasoning ( link to paper )
- The neuro-symbolic concept learner: Interpreting scenes, words, and sentences from natural supervision ( link to paper )
- Raven: A dataset for relational and analogical visual reasoning ( link to paper )
- Adaptive Input Representations for Neural Language Modeling ( link to paper )
- Understanding Grounded Language Learning Agents (link to paper )
- Mining for meaning: from vision to language through multiple network consensus (link to paper )
- On the Spectral Bias of Neural Networks (link to paper )
- Machine Learning Topological Invariants with Neural Networks (link to paper )
- Unsupervised discovery of demixed, low-dimensional neural dynamics across multiple timescales through tensor components analysis (link to paper )
- Stronger generalization bounds for deep nets via a compression approach (link to paper )
- Linear algebraic structure of word senses, with applications to polysemy ( link to paper )
- Gated orthogonal recurrent units: On learning to forget ( link to paper )
- A man field view of the landscape of a two-layer neural network (link to paper )
- Entropy and mutual information in models of deep neural networks (link to paper )
- … more to come …. (link to paper )