https://github.com/deshraj/toaim
It Takes Two to Tango: Towards Theory of AI's Mind
https://github.com/deshraj/toaim
ai beliefs humans intent mental-states mind theory
Last synced: about 2 months ago
JSON representation
It Takes Two to Tango: Towards Theory of AI's Mind
- Host: GitHub
- URL: https://github.com/deshraj/toaim
- Owner: deshraj
- License: mit
- Created: 2017-04-03T12:07:18.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2017-05-17T18:57:16.000Z (almost 8 years ago)
- Last Synced: 2025-01-07T19:22:58.251Z (4 months ago)
- Topics: ai, beliefs, humans, intent, mental-states, mind, theory
- Language: HTML
- Homepage: https://deshraj.github.io/TOAIM/
- Size: 46.4 MB
- Stars: 2
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# It Takes Two to Tango: Towards Theory of AI's Mind
## Authors
[Arjun Chandrashekharan](https://filebox.ece.vt.edu/~carjun/), [Deshraj Yadav](https://deshraj.github.io), [Prithvijit Chattopadhyay](https://prithv1.github.io), [Viraj Prabhu](https://virajprabhu.github.io/), [Devi Parikh](https://filebox.ece.vt.edu/~parikh/)
## Abstract
*Theory of Mind is the ability to attribute mental states (beliefs, intents, knowledge, perspectives, etc.) to others and recognize that these mental states may differ from one's own. Theory of Mind is critical to effective communication and to teams demonstrating higher collective performance. To effectively leverage the progress in Artificial Intelligence (AI) to make our lives more productive, it is important for humans and AI to work well together in a team. Traditionally, there has been much emphasis on research to make AI more accurate, and (to a lesser extent) on having it better understand human intentions, tendencies, beliefs, and contexts. The latter involves making AI more human-like and having it develop a theory of our minds.*
*In this work, we argue that for human-AI teams to be effective, humans must also develop a theory of AI's mind -- get to know its strengths, weaknesses, beliefs, and quirks.*
*We instantiate these ideas within the domain of Visual Question Answering (VQA). We find that using just a few examples (50), lay people can be trained to better predict responses and oncoming failures of a complex VQA model. Surprisingly, we find that having access to the model's internal states -- its confidence in its top-k predictions, explicit or implicit attention maps which highlight regions in the image (and words in the question) the model is looking at (and listening to) while answering a question about an image -- do not help people better predict its behavior.*
## Arxiv Link
https://arxiv.org/abs/1704.00717
## Citing this work
If you find this work useful in your research, please consider citing:
```
@article{toaim,
author = {Arjun Chandrasekaran and Deshraj Yadav and Prithvijit Chattopadhyay and Viraj Prabhu and Devi Parikh},
title = {{I}t {T}akes {T}wo to {T}ango: {T}owards {T}heory of {AI}'s {M}ind},
journal = {arXiv preprint arXiv:1704.00717},
year = {2017}
}
```
## License
[MIT](https://github.com/deshraj/TOAIM/blob/master/LICENSE)