Evaluations & project topics

This is a 6 credits course (If you want to take it for 9 credits please let me know and we can try to arrange something).

Evaluations

This course will have the following moments of evaluation:

Type	Percentage of grade	Info
Homework	30% (total) / 10% (each)	3 sets of homework. First set will be on the third week and then every two weeks for three times.
Final project	70%	Final project, either selected from the given list or proposed by you. (If the latter, please discuss your idea with me in advance)

The homework will be about the programming / technical side of the course.

Project topics

Here are some possible final projects in order of increasing complexity:

• Review of the literature on the probabilistic Language of Thought.

• Review of the philosophical literature on the Language of Thought.

• Implement a model for learning spatial connectives in LOTlib3 (e.g., ‘up’, ‘below’, etc.)

• Comment the source code of LOTlib3.

• Choose a Bayesian approximation algorithm (e.g. reversible-jump Monte Carlo), implement it, and fit one of the models discussed in the second half of the course.

• Implement a model of geometric shape learning in LOTlib3 (quite a challenge! Not sure how well it can be done, but I judge the quality of the attempt even if it does not work well).

• Choose your own semantic domain and implement a model that learns it in LOTlib3 (Discuss your choice of domain with me first!).