Tutorial on developing AI assistants

AI has recently shown impressive advances, from learning to play the Atari games to defeating expert human players in the game of Go. Beyond games, AI has also exploded in fields such as computer vision and natural language processing, where vast amounts of labeled data are available. More broadly speaking, technology as a whole has massively changed the landscape of most fields. However, current approaches can only help in tasks where we either can precisely specify the objective or already have plenty of observations of solutions to learn from.

However, important real-world problems rarely have well-specified objectives or solutions to learn from. Instead, most problems depend on the goals and preferences of humans - the users - who are solving them. As a result, we need approaches that explicitly consider the user. We, the Artificial agents with Theory Of Mind team of FCAI, do exactly that by developing techniques and methods that assist users in their tasks.

Here you can find a tutorial on creating AI-assistants.

Jupyter notebook setup Instructions

We recommend either:

1. Using Google Collab (if you have a Google account)
2. Setting up your own local environment

Instructions for Google Collab

Due to slight differences in how jupyter notebooks are rendered in Google Collab, you'll need to use the files located in the "Google Collab" folder.

1. Download or clone the repo.
2. Make sure you're signed into your Google account and open Google Collab
3. Navigate to the Upload tab > click choose file > go under the "Google Collab" directory > select the AI_assistance_Tutorial.ipynb file
4. Run the first cell in the notebook (it might take a bit of time for the Google backend to activate your virtual machine)
5. When the cell successfully runs, click choose file again and select the following two files for upload:
   • tutorialObjs.py
   • ai-assistance-overview.png
6. Run the subsequent import cell to verify you don't have any errors.

Instructions for local environment

Requirements

• Python 3 (we've tested on 3.8 and 3.9)
• jupyter notebook
• packages in requirements.txt

For pyenv or virtualenv

1. Download or clone the repo.

2. Instructions can differ depending on your tool, so if you need help, reference this external source

3. Once your environment is activated, remember to run pip install -r requirements.txt.

4. Run the jupyter notebook command

5. Open the AI_assistance_Tutorial.ipynb file and run the first cell to verify all imports were completed without error

For Anaconda environments

1. Download or clone the repo.

2. To creat a new environment named "myenv" with Python 3.8 installed, run the following command:

   conda create --name myenv python=3.8

3. After creating the environment, you need to activate it before installing any packages. Do this with the following command:

   conda activate myenv

4. Run the following command to install all the packages in requirements.txt:

   conda install --file requirements.txt

5. Run the jupyter notebook command

6. Open the AI_assistance_Tutorial.ipynb file from jupyter and run the first cell to verify all imports were completed without error