Skip to content
View Zajaczkowskim's full-sized avatar

Highlights

  • Pro

Block or report Zajaczkowskim

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
Zajaczkowskim/README.md

Hi 👋, I'm Michał Zajączkowski

About Me

Areas of intrest

I'm currently focused on building a solid foundation and gaining hands-on experience in the following areas:

  • Quantitative Research
  • Artificial Intelligence
  • Machine Learning
  • Data-Driven Problem Solving

Hackathon Projects

I've been involved in several hackathons over time, each one helping me grow my skills and learn something new. Here are a few events I’ve participated in:

  • HackYeah 20242nd place in the Activity and Wellness category
  • EnsembleAI4th place
    • Robustness to Adversarial Attacks - ResNet-based classifier trained to remain accurate on both clean and adversarially perturbed data using FGSM and PGD, optimizing the trade-off between robustness and standard accuracy.
    • Membership Inference Attack - Implemented a membership inference attack on a ResNet18 model, using shadow models and statistical techniques to assign confidence scores and maximize AUC ROC under limited data access.
    • Model Stealing Query-efficient model stealing pipeline to replicate a protected SSL encoder by minimizing the L2 distance between our surrogate model’s embeddings and the target’s outputs.

University projects

  • Bank Marketing Classification Model A supervised machine learning project focused on predicting the outcome of bank marketing campaigns based on client data using advanced preprocessing, feature engineering, and model evaluation techniques.
  • Soccer matches data analysis Clustering football players based on in-game performance statistics using unsupervised machine learning techniques.
  • Bioinformatics Recently discovered algorithms and developed computers give scientists chance to create more and more advanced models of chromatin that are very similar to the reality. This paper presents the result of the work on modelling chromatin using popular tools.
  • Messenger data visualisation dashboard developed in R (Shiny). The aim was to prepare interactive dashboard which presents data analysis about authors. We chose data from messaging app - Messenger and visualized them.
  • E-commerce store web application developed with Django, allowing users to browse, filter and purchase products.

Private projects

Tools

  • I have experience with:
    python java R pandas mysql scikit_learn

    git seaborn matlab

  • 🌱 Currently learning:
    tensorflow

Pinned Loading

  1. Financial-Mathematics Financial-Mathematics Public

    Jupyter Notebook

  2. Markov-Chains-Denoising-Images Markov-Chains-Denoising-Images Public

    Jupyter Notebook

  3. UserKrzysztof/MachineLearning-project1 UserKrzysztof/MachineLearning-project1 Public

    Jupyter Notebook 1

  4. Bioinformatics_Project Bioinformatics_Project Public

    Jupyter Notebook 2

  5. Soccer-matches-data-analysis Soccer-matches-data-analysis Public

    Forked from tomaszzywicki/Soccer-matches-data-analysis

    Jupyter Notebook

  6. projectMe projectMe Public

    Forked from Pacholki/projectMe

    Repository for our project for TWD

    CSS