ECINN: Efficient Counterfactuals from Invertible Neural Networks
Published at The British Machine Vision Conference (BMVC) 2021,
CV
Education
- Ph.D in Computer Science (Machine Learning), Aarhus University, December 2021
- M.S. in Computer Science, Aarhus University, 2018.
Specialization: Algorithms and Machine Learning.
Thesis: Explainable Deep Learning: Facilitating the Use of Deep Neural Networks in Healthcare, grade: 12.
Weighted GPA: 10.4 on a 12 scale
Courses from other universities: Advanced Machine Learning (DTU, grade: 12), Computational Data Analysis (DTU, 12), Algorithms for Massive Data Sets (DTU, 12), Introduction to Artificial Intelligence (UC Berkeley, A), Introductory Applied Econometrics (UC Berkeley, A). - B.S. in Computer Science, Aarhus University, 2015.
GPA: 8.91 on a 12 scale.
Bachelor thesis: Compiler for the programming language Tiger, Grade: 12.
Publications
Backpropagating through Fréchet Inception Distance
Published at arXiv,
What if Neural Networks had SVDs?
Published at NeurIPS 2020 (Spotlight),
One Reflection Suffice
Published at arXiv,
MeLIME: Meaningful Local Explanation for Machine Learning Models
Published at arXiv,
Faster Orthogonal Parameterization with Householder Matrices
Published at ICML Workshop on Invertyible Neural Networks, Normalizing Flows, and Explicit Likelihood Models,
Teaching
Pattern Recognition and Predictive Modelling in Neuroscience
Master Course at SDC, Neuroscience & Neuroimaging, Beijing, China
Data Mining
Master Course at AU, Dept. of Computer Science, Aarhus, Denmark
Introduction to Databases and Implementation and Applications of Databases
Bachelor Course at AU, Dept. of Computer Science, Aarhus, Denmark
Machine Learning
Bachelor Course at AU, Dept. of Computer Science, Aarhus, Denmark
Introduction to Databases
Bachelor Course at AU, Dept. of Computer Science, Aarhus, Denmark
Software Architecture
Bachelor Course at AU, Dept. of Computer Science, Aarhus, Denmark
Pervasive Computing (Operating Systems)
Bachelor Course at AU, Dept. of Computer Science, Aarhus, Denmark
Databases
Bachelor Course at AU, Dept. of Computer Science, Aarhus, Denmark
Work experience
- 2015-2016 Windows (WPF) Developer at Visiolink ApS, Aarhus, Denmark.
- 2013-2015 Student Developer at Visiolink ApS, Aarhus, Denmark
Talks
Fast Fréchet Inception Distance
Article Summary at Handout meeting, Aarhus University, Denmark
U-Days Presentation About Machine Learning
Introductory at U-Days, Aarhus University, Denmark
Neural Ordinary Differential Equations
Article Summary at Handout meeting, Aarhus University, Denmark
Presentation of PatternNet and PatternAttribution
Article Summary at Journal Club, Aarhus University, Denmark
AUHack - Tensorflow Introduction
Workshop at Major League Hacking - AUHack, Aarhus University, Denmark
Projects
Attention Mechanisms for Explanations
Many local post-hoc explainability techniques, such as DeConvNet, Guided Backprop, Layer-wise relevance propagation, and integrated gradients, rely on “gradient-like” computations, where explanations are propagated backwards through Neural Networks, one layer at a time. One can alter this backward computation to include attentions, which guides the explanation techniques to produce better explanations.
Skills
- Python
- Tensorflow
- Pytorch
- Keras
- Numpy
- MatLab
- R
- Java
- C++
Portfolio
- Layer-wise Relevance Propagation (LRP) framework for PyTorch 1.6 available on GitHub.
- Automatic Layer-wise Relevance Propagation (LRP) framework for Tensorflow 1.4 available on GitHub.
- Expectation Maximization for Kalman Filters GitHub.
- Material from study group on GANs GitHub.
Service and leadership
- 2016-2017 Board member of the Computer Science student organization (DSAU) as Aarhus University.
- Endless hours of gymnastics teaching.