Given the diversity of perspectives in recent artificial intelligence research, Nemodrive focuses on machine learning techniques, more specifically on developing algorithms that can generate a learning process for every type of sought output. In other words, an algorithm that can build on the initial data set and continuously learn how to drive on it’s own through experiences.
- Can we enhance training by assisting it with a symbolic representation ?
- Can we bootstrap and improve the learned policy through real human experiences ?
- How much interpretability and explainability can we achieve for the self-driving „brain”?