Name: Christian D. Glissov
Type: User
Bio: "...only the last line is a genuine theorem here - everything else is in the fantasy."
I love surfing the latent space.
https://cdglissov.github.io/
Location: Denmark, Copenhagen
Christian D. Glissov's Projects
Exercises and supplementary material for the deep learning course 02456 using PyTorch.
Projects with statistical modelling (GLMs, likelihood, HMM, mixtures, mixed effect models)
We set out to investigate how very simple data augmentations, applied at test time, can be used to defend against adversarial attacks.
Bayesian Data Analysis course at Aalto
Using Reinforcement learning to play Bomberman
Front-end fun. Created using HTML, Bootstrap, JavaScript, Sass, GitHub actions, and Webpack.
Work done in the course 02807
Exercises with constrained optimization (SQP, Trust regions, Newton, LP, Markowitz, KKT)
Play with some css
Projects in the course 02807
Dotfiles backup
Small Java project
Fun with some computer vision projects (CycleGAN, segmentation, sliding window)
This will be a future repo of different GANs
Creating graph data and comparing linear and geometric approaches
🦜🔗 Build context-aware reasoning applications
LaTeX.css is a CSS library that makes your website look like a LaTeX document
Using dynamic LDA to extract topics from speeches in the danish parliament.
Testing a t-distribution assumption for the latent space and decoder. The model is from this paper: https://arxiv.org/abs/1812.02633
Analysis of the correlation between demographical indicators and spread of COVID-19.
Just experimenting with dbt.
Combining normalizing flows and deep latent variable models to generate video.
Simple research database with SQL.
Assignments done in the course social data analysis
Project about supply chains and predictability between markets.
Projects with time series analysis (SETAR, IGAR, MMAR, MARIMA, CTSM, Kalman, GARCH, RLS)