Karolis Martinkus

Distributed Computing Group
Computer Engineering and Networks Laboratory (TIK)
Department Electrical Engineering (D-ITET)
Office ETZ G 60.1
ETH Zurich

Gloriastrasse 35
8092 Zurich
Switzerland

phone	+41 44 63 20429
fax	+41 44 63 21036
email

Publications

Discovering Graph Generation Algorithms
Mihai Babiac, Karolis Martinkus and Roger Wattenhofer.
Neurosymbolic Generative Models Workshop (NeSy-GeMs @ ICLR 2023), Kigali, Rwanda, May 2023.
BibTeX
Agent-Based Graph Neural Networks
Karolis Martinkus, Pál András Papp, Benedikt Schesch and Roger Wattenhofer.
11th International Conference on Learning Representations (ICLR), Kigali, Rwanda, May 2023.
BibTeX
Diffusion Models for Graphs Benefit From Discrete State Spaces (Extended Abstract)
Kilian Konstantin Haefeli, Karolis Martinkus, Nathanaël Perraudin and Roger Wattenhofer.
First Learning on Graphs Conference (LoG 2022), Virtual, December 2022.
External BibTeX
Automating Rigid Origami Design
Jeremia Geiger, Karolis Martinkus, Oliver Richter and Roger Wattenhofer.
ICML Workshop on Machine Learning for Computational Design (MLCD), Baltimore, Maryland, USA, July 2022.
BibTeX
SPECTRE: Spectral Conditioning Overcomes the Expressivity Limits of One-shot Graph Generators
Karolis Martinkus, Andreas Loukas, Nathanaël Perraudin and Roger Wattenhofer.
39th International Conference on Machine Learning (ICML), Baltimore, Maryland, USA, July 2022.
BibTeX
On Isotropy Calibration of Transformers
Yue Ding, Karolis Martinkus, Damian Pascual, Simon Clematide and Roger Wattenhofer.
3rd Workshop on Insights from Negative Results in NLP (Insights @ ACL 2022), Dublin, Ireland, May 2022.
External BibTeX
DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks (Spotlight)
Pál András Papp, Karolis Martinkus, Lukas Faber and Roger Wattenhofer.
35th Conference on Neural Information Processing Systems (NeurIPS), Virtual, December 2021.
BibTeX
Scalable Graph Networks for Particle Simulations
Karolis Martinkus, Aurelien Lucchi and Nathanaël Perraudin.
35th AAAI Conference on Artificial Intelligence (AAAI), Virtual, February 2021.
BibTeX

Theses & Labs

	Title	Type	Contact/Supervisor	Assigned	Student(s)
Current Theses & Labs
	How Robust Are the Expressive Graph Neural Networks?	B	Karolis Martinkus	FS 23
	Spectral Coarsening for Graph Generation	M	Karolis Martinkus	FS 23
Past Theses & Labs
	Title	Type	Contact/Supervisor	Assigned	Student(s)
	Contrastive Learning using SPECTRE for Generating Graph Augmentations	S	Karolis Martinkus, Zhao Meng	HS 22
	Developing Interpretable Graph Neural Networks for High Dimensional Feature Spaces	B	Karolis Martinkus, Lukas Faber	HS 22
	Node-level Prediction Tasks with Agent-based Graph Neural Networks	B	Karolis Martinkus	HS 22
	Spectral Denoising Diffusion Probabilistic Models for Graphs [confidential]	S	Karolis Martinkus	HS 22
	Discrete Denoising Diffusion Probabilistic Models for Graphs [confidential]	B	Karolis Martinkus	FS 22
	Vector-Quantized Generative Adversarial Networks for Graphs	M	Karolis Martinkus	FS 22
	Discovering Graph Generation Algorithms [confidential]	M	Karolis Martinkus	FS 22
	Agent-based Graph Neural Network [confidential]	S	Karolis Martinkus	HS 21
	Distributed Asynchronous Policy (DAP) with RL [confidential]	M	Karolis Martinkus, Lukas Faber	HS 21
	CloudMLS: A Cloud-based E2EE scheme	M	Karolis Martinkus, Ard Kastrati	HS 21
	Interpretable Graph Neural Network [confidential]	M	Karolis Martinkus, Lukas Faber	HS 21
	Scaling Attention Using Binary Trees [confidential]	M	Karolis Martinkus	HS 21
	Distance Preserving Graph Embedding	B	Karolis Martinkus	FS 21
	End to End Encryption in a Cloud-Based Messenger [confidential]	S	Ard Kastrati, Karolis Martinkus	FS 21
	Graph Coloring with Deep Learning [confidential]	S	Karolis Martinkus	FS 21
	Enhancing Graph Neural Networks with Boosting	B	Lukas Faber, Karolis Martinkus	FS 21
	Neural Distance Oracle for Road Graphs	B	Karolis Martinkus	FS 21
	On Isotropy Calibration of Transformer Models [confidential]	M	Damian Pascual, Karolis Martinkus	HS 20
	A Rigid Origami Environment for Reinforcement Learning [confidential]	M	Oliver Richter, Karolis Martinkus	HS 20

Karolis Martinkus

Publications

Theses & Labs

Current Theses & Labs

Past Theses & Labs