Active Learning of Conditional Mean Embeddings via Bayesian Optimisation

Code repository for the paper:

Chowdhury, Sayak Ray, Rafael Oliveira, and Fabio Ramos. 2020. “Active Learning of Conditional Mean Embeddings via Bayesian Optimisation.” In Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI). PMLR volume 124.

Available at:

• Paper: http://auai.org/uai2020/proceedings/469_main_paper.pdf
• Supplement: http://auai.org/uai2020/proceedings/469_supp.pdf

Requirements

The code is based on Python 3.7. To run the code, we recommend creating a Python 3.7 virtual environment, with e.g. Anaconda, and then installing the requirements within the virtual environment:

pip install -r requirements.txt

Toy experiment

You can run the toy experiment in the paper with:

python toy_experiment.py

Plots

A Jupyter notebook plotting.ipynb is provided with code to plot the optimisation performance. The notebook loads the output files of toy_experiment.py , which are named as <method_name>-regret.pth and saved to the current directory. For convenience, the regret data files used to generate the plots in the paper are provided, so that the following plot can be reproduced with the notebook.