Azamat Omuraliev's Projects
repo for solving advent of code
Encoding Social Choice theorems in propositional logic to obtain automated proofs.
My README!
Analyze your conversations with ChatGPT to find interesting patterns and actionable learnings about your interests and your career growth.
Telegram bot for budget planning and personal finance management
Fine-tune an LLM on email subjects to create better content calendars.
Repository for Project 3A: "Jointly learning natural language inference and fake news detection" of the SMNLS course taught at the UvA in the Spring 2019 semester
Transfer learning on SNLI-trained encoders
Lab 3 at UVA covering Kernel Methods
Machine Learning 2 Labs
Master Thesis
This repository holds all code for our multi-modal Fake News Detection model, which combines a GCN and a LM. This model was designed and implemented in the context of the Project AI course taught at the UvA in the Spring 2019 semester
A repository for sharing and collaboration for third-party Mycroft skills development.
An experiment on power analysis in information retrieval. In this notebook, we simulate click models to compare two ranking models (A/B testing) in online environment given offline metrics. Click models implemented: Random Click Model, Position-Based Model. Interleaving methods implemented: Team-Draft, Probabilistic.
š | A simple worker that can be used as a starting point to build your own custom RunPod Endpoint API worker.
KR, UvA, MSc
Slides from talk at Datafest Tbilisi 2023 "Using data science to discover problems worth solving"
Slides from talk at PyData Amsterdam "Building true machine learning MVPs"
Notebook exploring how unsupervised learning works for categorical, one-hot encoded features.
This repository contains implementations of algorithms proposed in recent papers from top machine learning conferences on Fairness, Accountability, Confidentiality, and Transparency (FACT). These were implemented as part of the FACT-AI course at the University of Amsterdam.
Summarize YouTube videos in one click!