This project involves two sequential tasks aimed at understanding a dataset's underlying structure and behavior. The dataset contains 12 independent variables and a single dependent variable $y$ such that $y = f(x_1, x_2, \dots, x_{12})$. This is incorporated in the csv file in this folder.

• Task 1: Optimize the function $f(x)$ to find its minima using any suitable method.
• Task 2: Represent the functional relationship between $y$ and the variables $(x_1, x_2, \dots, x_{12})$ in a comprehensible form. This involves employing methods that enhance explainability in artificial intelligence (AI).

Find the minima of $f(x_1, x_2, \dots, x_{12})$

• Use any method or approach you are comfortable with, such as gradient descent, genetic algorithms, or other optimization techniques.

Represent the relationship $y = f(x_1, x_2, \dots, x_{12})$ in an explainable manner.

• Develop a mathematical equation or logical model that captures the functional relationship between the variables.
• The focus should be on generating a form that is interpretable by humans, such as a regression equation, tree structure, or rule-based model.
• Explainable AI (XAI) methods should be emphasized to ensure that the results are not just accurate but also transparent and understandable.

Explainable AI ensures that the models are interpretable and their predictions are trustworthy. By creating a functional representation, this task aims to bridge the gap between black-box models and human understanding.

Read other papers which discuss the explanability of black box models.

Example : Survey

Fork this repo, and add your notebook to the solutions folder as "your_name.py"