~mikelma
Hi! My name is Mikel Malagón,
mikelma
online. I received a degree in Computer Science from the
University of the Basque Country UPV/EHU
in 2022, and a master's degree in Computer Engineering and
Artificial Intelligence the following year from the same
university. Currently, I am a PhD student in the
Intelligent Systems Group
under the supervision of
Josu Ceberio
and
Jose A. Lozano.
My main research interests are (continual) reinforcement
learning, open-ended learning, and deep learning. Although
my research mainly focuses on ML, I'm also very interested
in compilers, programming languages, and free software.
Besides science and technology, I love rock climbing and
playing the guitar.
Contact: mikel DOT malagon AT
ehu DOT eus
✈️ Upcoming travel
I’ll be traveling to Canada 🇨🇦 from March to May as a visitor researcher in the RLAI lab at Univerisity of Alberta with Prof. Michael Bowling.
News
Sep 03, 2025
I'll be attending to
ECAI
in Bologna 🇮🇹 during late october.
Jun 29, 2025
Craftium
has reached over 100 stars ⭐ on GitHub! Thanks to all
stargazers!
Jun 25, 2025
I'll be presenting
"Craftium: Bridging Flexibility and Efficiency for Rich
3D Single- and Multi-Agent Environments"
as an oral in the
RLVG Workshop
at
RLC 2025
in Edmonton 🇨🇦 this August.
May 23, 2025
I'll be attending the
DLRL Summer School
in Edmonton 🇨🇦 this July.
May 02, 2025
I'll present
"Craftium: Bridging Flexibility and Efficiency for Rich
3D Single- and Multi-Agent Environments"
at ICML on July 17. See you in Vancouver! 🇨🇦
May 02, 2024
I'll be presenting
"Self-Composing Policies for Scalable Continual
Reinforcement Learning"
as an oral presentation at ICML 2024 in Vienna! 🇦🇹
Selected publications
[ICML 2025]
(🏅 @ RLVG)
(🏅 @ RLVG)
"Craftium: Bridging Flexibility and Efficiency for Rich
3D Single- and Multi-Agent Environments"
Mikel Malagón, Josu Ceberio, Jose A. Lozano
Published at ICML, also presented as an oral 🗣️ in the RLVG Workshop @ RLC 2025 [RLVG Best Paper Award 🏅]
Mikel Malagón, Josu Ceberio, Jose A. Lozano
Published at ICML, also presented as an oral 🗣️ in the RLVG Workshop @ RLC 2025 [RLVG Best Paper Award 🏅]
[ICML 2024]
(Oral 🗣️)
(Oral 🗣️)
"Self-Composing Policies for Scalable Continual
Reinforcement Learning"
Mikel Malagón, Josu Ceberio, Jose A. Lozano
Mikel Malagón, Josu Ceberio, Jose A. Lozano
Projects
A non-exhausistive list of my projects (excluding papers):nan-audio
NaN
(Not a Noise) is a real-time audio syntheis extension for GNU
Guile designed for live coding.
NiGUI
Real-time EEG data visualization UI. This software is part of the NIT project @ Medialabs (Tabakalera). The project has been developed in collaboration with the
Basque Center On Cognition Brain and Language
.
radio-CARLOS
The amateur radio telescope of the astronomy group @ Medialabs
(Tabakalera). Code
here
.
okta-lang
Compiled, imperative, and general-purpose programming language that
aims to provide a simple platform to create efficient software.
Main repo
here
.