Thanks to visit codestin.com
Credit goes to github.com

Skip to content
View mirkzx04's full-sized avatar

Highlights

  • Pro

Block or report mirkzx04

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mirkzx04/README.md

Hi, I'm Mirko

I’m a student at Sapienza University of Rome, studying Mathematical Sciences for Artificial Intelligence.

I’m genuinely passionate about Deep Learning not just using models, but rebuilding them from scratch to understand every detail. I like experimenting with architectures, testing different hyperparameters, and playing with optimization techniques to see how they shape a model’s behavior.

I keep up with the field by reading a lot of research papers, from classic works to the latest releases. I enjoy taking these theoretical ideas and trying to apply them to real problems or to my own projects, like the PCE experiment I’m working on.

Right now, I’m especially interested in Generative AI and the fast progress of Large Language Models. Their emergent, almost human-like behaviors fascinate me not just their strengths, but also their quirky limitations, like sycophancy, which I find particularly interesting to study.

🛠️ My Tech Stack

Python PyTorch TensorFlow Git

📫 How to reach me

LinkedIn

Pinned Loading

  1. Positional_Convolution_Experts Positional_Convolution_Experts Public

    This is a project for my Machine Learning exam. I decided to try a new architecture that I thought up, the architecture consists of a CNN where in each layer there is a positional router, the convo…

    Python 2

  2. n-bandits-problem n-bandits-problem Public

    A Python implementation of the Gradient Bandit algorithm using a softmax policy with baseline.

    Python

  3. Ridge-Regression-with-QR-Ortomin-Cnjugate-Gradient-and-SDG Ridge-Regression-with-QR-Ortomin-Cnjugate-Gradient-and-SDG Public

    Implementation of Ridge Regression with QR, Orthomin, and Conjugate Gradient

    MATLAB