https://github.com/jonathan-cook235/cultural-accumulation

https://github.com/jonathan-cook235/cultural-accumulation

Last synced: 5 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/jonathan-cook235/cultural-accumulation
• Owner: jonathan-cook235
• Created: 2024-05-15T08:53:19.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2024-06-11T15:33:59.000Z (5 months ago)
• Last Synced: 2024-06-11T23:08:18.984Z (5 months ago)
• Language: Jupyter Notebook
• Size: 70.3 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Cultural Accumulation in Reinforcement Learning

Code for replicating experiments run in _Artificial Generational Intelligence: Cultural Accumulation in Reinforcement Learning_.

## Memory Sequence Experiments

To run these experiments, simply open the Jupyter notebook __memory_sequence.ipynb__ and step through the cells.

## TSP Experiments

To run these experiments, simply open the Jupyter notebook __tsp.ipynb__ and step through the cells.

## Goal Sequence Experiments

### In-Context Accumulation

1. To train an oracle, run __goal_seq/train_oracle_s5.py__.

2. To train a social learner, run __goal_seq/train_social_s5.py__.

3. To evaluate in-context accumulation, run __goal_seq/in_context_accumulation.py__.

### In-Weights Accumulation

For this experiment, run __goal_seq/in_weights_accumulation.py__.