Software

Research software

dtaianomaly

Time series anomaly detection is a crucial task in many applications, and decades of research lead to hundreds of algorithms. ´dtaianomaly´ is an easy-to-use Python package that provides implementations to a large collection of algorithms, ranging from classical nearest-neighbour approaches to the most recent, cutting-edge developments.

PatsEmb

The pattern-based embedding is an embedding of the subsequences in a time series which indicates which frequent patterns occur in the subsequence. We have developed this method for semantic segmentation, but is applicable on many other time series analysis tasks.

Heuristic algorithms using tree decompositions

Tree decompositions are typically used to develop exact, polynomial time algorithms for NP-hard graph problems for graphs of bounded tree width. We adapted these algorithms to a heuristic solution and applied the method to the Maximum Happy Vertices Problem.

Side projects

Human Benchmark

The Human Benchmark measures the human abilities through a set of brain games and cognitive tests. The goal of this project is to develop methods to beat these games and achieve the highest possible scores.

MathViz

A simple Python project designed to bring mathematical patterns to life through interactive visualizations.

Conway's Game of Life

A simulation of Conway's Game of Life. This is a cellular automata consisting of dead and alive cells, in which the number of alive neighbours of a cell decide whether a cell will be alive in the next step.