Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/thesofakillers/nlgoals
Official repository for my MSc thesis: "Addressing Goal Misgeneralization with Natural Language Interfaces."
https://github.com/thesofakillers/nlgoals
ai-safety babyai foundation-models goal-misgeneralization llm nlp rl
Last synced: about 5 hours ago
JSON representation
Official repository for my MSc thesis: "Addressing Goal Misgeneralization with Natural Language Interfaces."
- Host: GitHub
- URL: https://github.com/thesofakillers/nlgoals
- Owner: thesofakillers
- Created: 2022-10-14T15:25:53.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2023-11-02T22:10:34.000Z (about 1 year ago)
- Last Synced: 2023-11-07T14:24:27.543Z (about 1 year ago)
- Topics: ai-safety, babyai, foundation-models, goal-misgeneralization, llm, nlp, rl
- Language: TeX
- Homepage:
- Size: 35.7 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Addressing Goal Misgeneralization with Natural Language Interfaces
Official repository for my MSc thesis in Artificial Intelligence at the
University of Amsterdam:
_Addressing Goal Misgeneralization with Natural Language Interfaces_
Find and read the thesis
[here](https://scripties.uba.uva.nl/search?id=record_53979;setlang=en) or
[here](./documents/thesis-main/main.pdf).
## Requirements and Setup
Details such as python and package versions can be found in the generated
[pyproject.toml](pyproject.toml) and [poetry.lock](poetry.lock) files.
We recommend using an environment manager such as
[conda](https://docs.conda.io/en/latest/). After setting up your environment
with the correct python version, please proceed with the installation of the
required packages
First, install the pre-requirements[^1]:
```terminal
pip install -r requirements/pre-reqs.txt
```
Then install the rest of the requirements:
```terminal
pip install -r requirements/requirements-complete.txt
```
These `requirements.txt` file are generated by running
[gen_pip_reqs.sh](gen_pip_reqs.sh)
[^1]:
This annoying first step is necessary because of stupid packages we depend
on like pyhash and multicoretsne that don't know how to get their
dependencies sorted out like everyone else. See
[this issue](https://github.com/DmitryUlyanov/Multicore-TSNE/issues/81#issuecomment-863745998)
and [this issue](https://github.com/flier/pyfasthash/issues/59). These two
packages are the main culprits but other packages haven't been very good at
defining their dependencies either.
## Project Organization
```plaintext
├── LICENSE
├── README.md <- The top-level README
├── data/ <- Datasets
├── checkpoints/ <- Trained and serialized models.
├── notebooks/ <- Jupyter notebooks.
├── documents/ <- Documents as HTML, PDF, LaTeX, etc.
├── pyproject.toml <- Project metadata, handled by poetry.
├── poetry.lock <- Resolving and locking dependencies, handled by poetry.
├── requirements.txt <- For non-poetry users.
├── gen_pip_reqs.sh <- For generating the pip requirements.txt file
├── tests/ <- Tests
├── outputs/ <- Output files. Not committed.
└── src/nlgoals/ <- Source code for use in this project.
├── __init__.py <- Makes src a Python module
├── babyai/ <- (modified) code from babyai repo
├── configs/ <- Miscellaneaous configuration
├── utils.py <- Miscellaneaous utils
├── data/ <- Data processing and handling
├── interfaces/ <- Logic for connecting datasets and models
├── models/ <- Model definitions
├── losses/ <- Custom loss function implementations
├── trainer/ <- Configurations for training
└── run/ <- Scripts to train, evaluate and use models
```
The project structure is largely based on the
[cookiecutter data-science template](https://github.com/drivendata/cookiecutter-data-science).
This is purposely opinionated so that paths align over collaborators without
having to edit config files. Users may find the
[cookiecutter data-science opinions page](http://drivendata.github.io/cookiecutter-data-science/#opinions),
of relevance
The top level `data/` and `checkpoints/` directories are in version control only
to show structure. Their contents will not be committed and are ignored via
`.gitignore`.
## Demo videos
"turn on the light bulb"
https://github.com/thesofakillers/nlgoals/assets/26286291/e6849007-11f0-44e9-9d66-3b2b575492fc
"go to the key"
https://github.com/thesofakillers/nlgoals/assets/26286291/4d8711f4-e53a-4c7f-bd2d-9fb5a2ee1bdb
---