https://github.com/schizyfos/Goal
Creating a community for turning Valiant's and Rényi's vision into an AI model
https://github.com/schizyfos/Goal
ai ethics learnability
Last synced: 3 months ago
JSON representation
Creating a community for turning Valiant's and Rényi's vision into an AI model
- Host: GitHub
- URL: https://github.com/schizyfos/Goal
- Owner: schizyfos
- Created: 2021-12-15T05:19:29.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2024-11-25T16:02:37.000Z (about 1 year ago)
- Last Synced: 2025-01-06T00:35:31.826Z (11 months ago)
- Topics: ai, ethics, learnability
- Homepage:
- Size: 26.4 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome_ai_agents - Goal - Creating a community for turning Valiant's and Rényi's vision into an AI model (Building / Ethics)
README
# Goal
Creating a community for the development of AI learning model, outlined in the references bellow.
# References
[1] Probably approximately correct (PAC), by Leslie Valiant,
https://people.mpi-inf.mpg.de/~mehlhorn/SeminarEvolvability/ValiantLearnable.pdf
"The discovery of the notion of computability constituted a new approach to discovering truths about the world. The logician Kurt Gödel generously acknowledged that computability theory has for the first time succeeded in giving an absolute definition of an interesting epistemological notion, i. e., one not depending on the formalism chosen. What can be computed does not change as one varies the details of the model."
[2] A Socratic dialogue on mathematics, by Alfréd Rényi,
https://www.cambridge.org/core/services/aop-cambridge-core/content/view/13265932B04BE25A51D3944209C3AA08/S0008439500052346a.pdf/socratic_dialogue_on_mathematics.pdf
"Why do you think, Socrates, that this method of thinking and arguing can be used only for the study of numbers and geometric forms? Why do you not try to convince your fellow-citizens to apply the same high logical standards in every other field, for instance in philosophy and politics; in discussing the problems of everyday private and public life?" From that time, this is what I have always tried to do".
[3] PAC learning and Deep nets
https://elmos.scripts.mit.edu/mathofdeeplearning/2017/05/08/mathematics-of-deep-learning-lecture-4/
[4] PAC learning in the presence of evasion adversaries
https://www.princeton.edu/~pmittal/publications/pac-neurips18.pdf
[5] On the universality of deep learning
https://proceedings.neurips.cc//paper/2020/file/e7e8f8e5982b3298c8addedf6811d500-Paper.pdf
[6] PAC confidence sets for deep neural networks via calibrated prediction
https://openreview.net/pdf?id=BJxVI04YvB
[7] Learning algorithms from circuit lower bounds
https://arxiv.org/abs/2012.14095
[8] Learning algorithms versus automatability of Frege systems
https://arxiv.org/abs/2111.10626
# Bratislava circle
An international creative circle, (inspired by Vienna circle from the beginning of the last century) of theorists and practitioners to achieve this top noble and interesting goal.