Welcome to the repository of GDonut! This program is designed to help students visually explore the method for constructing non-expansive equivariant group operators (GENEOs) through the input of permutants and symmetrical functions.
A GENEO, or non-expansive equivariant group operator, is a mathematical construct that can be applied to construct a new paradigm of learning networks, based on the formal study of the topological space of these operators.
Equivariance refers to the property of a function or operator that remains unchanged when applied to a transformed version of its input. In other words, if a function or operator is equivariant, it will give the same output no matter how its input is transformed.
Non-expansiveness, on the other hand, refers to the property of a function or operator that does not increase the distance between its inputs. This property is important for ensuring that the output of the function or operator is not significantly different from its input, which can be useful for tasks such as learning or optimization.
GENEOs are useful in the construction of learning networks, as they can be used to create models that are robust to transformations of their input data. They are also useful for studying the topological space of group operators, which can provide insights into the structure of data and the relationships between different data points.
References
Non-Expansive Equivariant Group Operators for Artificial Intelligence
program Features
Select groups and permutants
Insert symmetrical functions
Process images from input data
Save and reuse the GENEO as an image for further studies
Normalize the GENEO obtained by an arbitrary constant
Technologies Used
Tauri
Rust
Vue.js
Vite.js
[ml-matrix]([url](https://mljs.github.io/matrix/))
Contributions
We welcome contributions to the program. If you have any suggestions or find any bugs, please open an issue or submit a pull request.
Thank you for using the GDount program!