About

Rediger foto

I'm an ELLIS PhD Student at the University of Copenhagen, advised by Prof & Director Serge Belongie. Additionally I'm also part of the Pioneer Centre for Artificial Intelligence and BelongieLab.

Before starting my PhD, I was a visiting student at UC Berkeley and a research intern at RISELab, where I was fortunate to work with Xinyun Chen in Dawn Song's group.

I received my Bachelor's and Master's degrees from the Technical University of Denmark DTU (DTU), where I worked with Prof. Ole Winther Winther and Prof. Morten Mørup. I've also previously been at Raffle.ai , a TA in DTU's Deep Learning course 02456 and worked a Machine Learning Engineer at Corti.

Research Interest

I am broadly interested in research on narratives and misinformation in NLP. I believe that the time is ripe for creating new takes on current misinformation pipelines that are more flexible and can handle any textual data. I broadly work in the HCI+NLP+CV area, but I specialise in topics such as argument mining and misinformation, as well as text summarization and style transfer.

peter_ebert@live.dk GitHub LinkedIn Resume

Publications

Searching for Structure in Unfalsifiable Claims
Peter Ebert Christensen, Frederik Warburg, Menglin Jia, Serge Belongie
github, 2022
PDF Abstract Bibtex Code

Volumetric Disentanglement for 3D Scene Manipulation
Sagie Benaim, Frederik Warburg, Peter Ebert Christensen, Serge Belongie
arxiv, 2022
PDF Abstract Bibtex Code

Synthesize, Execute and Debug, Learning to Repairfor Neural Program Synthesis
Kavi Gupta, Peter Ebert Christensen, Xinyun Chen, Dawn Song
NeurIPS 2020, 2020
PDF Abstract Bibtex Code

A Deep Learning Approach to Short Term Blood Glucose Prediction based on Continuous Glucose Monitoring Data
Ali Mohebbi, Alexander Johansen, Nicklas Hansen, Peter Ebert Christensen, Morten Mørup
IEEE EMBC, 2020
PDF Abstract Bibtex Code

Autoencoding undirected molecular graphs with neural networks
Jeppe Olsen, Peter Eber Christensen, Martin Hansen, Alexander Rosenberg Johansen
arxiv, 2019
PDF Abstract Bibtex Code


Teaching

Below you can find some of the material I used for courses and workshops where I have teached.

DTU course 02456 Deep learning
Programming Exercises (PyTorch) for the Deep Learning Graduate Course at the Technical University of Denmark running in the Fall of 2016
Code

Neural AI
Explaining Reinforcement Learning for more than 200 people at the Technical University of Denmark during a Neural AI event
Code


Projects

Below you can find a list over projects and Internships I've been doing so far

DTU course 02456 Deep learning
Teaching assistant, designed programming exercises in PyTorch and supervising over 40 students in their group projects
Fall, 2019
Code Website

Cs 294-158 Deep Unsupervised Learning UC Berkeley (phD level course)
I have implemented archtechtures and techniques from about 10 papers, been updated with the newest techniques within generative modelling and unsupervised learning by covering more than 100 papers through reading papers and following lectures from renowned professors and ph.D students from UC Berkeley
Fall, 2019
Solutions Website

Internship at Raffle.ai
Machine learning Intern, build text2sql models
July, 2019
Code Website

DTU special course Deep reinforcement learning
Implementing an efficient multiscale A3C algorithm with advantage estimation as well as completing the other half of UC Berkeley's Deep RL course
June, 2019
Code Website

DTU special course Introduction to reinforcement learning
Reading and implementing classical TD, Reinforce, Dyna-Q, DDQN from the Barto Sutton book and completing the first half the Deep RL cource from UC Berkeley
Spring, 2019
Code Solutions Book

DTU course 02456 Deep learning
Implement an RL agent using PPO and improving its exploration capabilities with Random Network disillation for the pommerman challenge, which was sponsored by NVidia and Google
Fall, 2018
Code Challenge

Student assistant / Bachelor project
Helped with building the core infrastructure in Python for the Autonomous Materials discovery and implementing several machine learning models for automated analysis, headed by Prof. Tejs Vegge
Fall, 2018
Website

DTU course 02460 Advanced Machine Learning
Implemented and developed Nonnegative matrix factorization methods trained with variational inference during a research project headed by Mikkel N. schmidt
Spring, 2018
Code google scholar