Natural Language Understanding assignment 1 about tasks to understand Recurrent Neural Networks. An analysis report was written based on the questions from the description file.
- Implementation of a few critical parts of a recurrent neural network and the back-propagation algorithm.
- Experiment in training to adapt the model to a task about predicting number agreement between subject and predicate in English.
- Implementation of RNNs and backpropagation.
- Parameter tuning in training language models.
- Analyzed effects and chose from a set of parameters.