Pinned Repositories
g2-lewis
g3-nandi
beta-vae-group2
constraint-assistance-games
A repository for solving Constraint Assistance Games and Ethical Compliance Assistance games: variants of the Assistance Game / Cooperative Inverse Reinforcement Learning framework, focussed on learning constraints by playing a game with a human.
RL-DIM
C. Griffin, J. Skalse, L. Hammond, and A. Abate, “All’s well that ends well: Avoiding side effects with distance-impact penalties”, Accepted at NeurIPS 2022 ML Safety Workshop, 2022.
lmorl
Lexicographic Multi-Objective Reinforcement Learning
al-llm-experiments
Experiments on using a generative text model with active learning to train a text classifier
wildfire
CJ-Griffin's Repositories
CJ-Griffin/RL-DIM
C. Griffin, J. Skalse, L. Hammond, and A. Abate, “All’s well that ends well: Avoiding side effects with distance-impact penalties”, Accepted at NeurIPS 2022 ML Safety Workshop, 2022.
CJ-Griffin/beta-vae-group2
CJ-Griffin/constraint-assistance-games
A repository for solving Constraint Assistance Games and Ethical Compliance Assistance games: variants of the Assistance Game / Cooperative Inverse Reinforcement Learning framework, focussed on learning constraints by playing a game with a human.