Silvia Arellano García
I'm
About
I am a recent graduate in Machine Learning from KTH Royal Institute of Technology (Stockholm, Sweden). Previously, I earned
a BSc in Electrical Engineering from the Universitat Politècnica de Catalunya (Barcelona, Spain). I have gained research experience in
both industry (Dolby Laboratories, Electronic Arts) and academia.
Currently, I enjoy working on topics related with Deep Learning and Generative AI, with a particular interest in the field of
animation. I like the idea of working on the intersection of art and technology! Additionally, I am a big fan of animated films,
it would be so cool to be able to contribute in some way to their creation.
Feel free to contact me via email or LinkedIn if you would like to discuss research topics related to my interests or explore
potential opportunities. I would be open to consider relocating if necessary.
Education & Recent positions
Education
MSc Machine Learning
2023 - 2025
KTH Royal Institute of Technology, Stockholm, Sweden
Relevant subjects: Advanced Deep Learning, Advanced Machine Learning, Computer Graphics and Visualization, Computer Vision, AI for Multiagent Systems
BSc Electrical Engineering
2019 - 2023
ETSETB - Universitat Politècnica de Catalunya (UPC), Barcelona, Spain
Specialized in Audiovisual Systems. Relevant courses: Image and Video Processing, Audio and Speech Processing, Probability and Statistics, Audiovisual Coding, Acoustics and Electroacustics.
Recent positions
ML Engineer @ EA Sports
September 2025 - Present
Part of the Skeleton Team at TRACAB, working on computer vision problems.
Master Thesis Intern @ Electronic Arts (SEED)
Spring 2025
Supervised by Héctor Anadón León and Judith Bütepage
Generation of 3D assets from multi-view images using flow-matching models. Used TRELLIS to develop new latent space conditioning strategies that enhance the consistency of geometry across views.
Research intern in AI and acoustics @ Dolby Laboratories
Summer 2024
Supervised by Daniel Arteaga
Interned at the Advenced Technology Group Department. Conducted research on generating Room Impulse Responses using deep neural networks conditioned on acoustic parameters, adapting models like autoregressive transformers, flow matching, and MaskGIT.
Teacher Assistant @ KTH Royal Institute of Technology
Spring and Autumn semesters 2024
Courses: DD2257 Visualization (Period 1) and DD1420 Foundations of Machine Learning (Period 3). Responsible for grading assignments and conducting help sessions.
Undergraduate Research Fellow @ UPC Computer Architecture department
2022 - 2023
Supervised by Beatriz Otero and Rubèn Tous
Explored the use of Diffusion Models and Generative Adversarial Networks (GANs) for digital graphics creation. Funded by a competitive scholarship from the Spanish Ministry of Education.
Publications
Room Impulse Response Generation Conditioned on Acoustic Parameters
Silvia Arellano, Chunghsin Yeh, Gautam Bhattacharya and Daniel Arteaga.
WASPAA 2025 - IEEE Workshop on Applications of Signal Processing to Audio and Acoustics
A 3D terrain generator: Enhancing robotics simulations with GANs
Silvia Arellano, Beatriz Otero, Tomasz Piotr Kucner, and Ramon Canal.
LOD2023 - International Conference on Machine Learning, Optimization, and Data Science
Exploring Image Transformations with Diffusion Models: A Survey of Applications and Implementation Code
Silvia Arellano, Beatriz Otero, Ruben Tous.
LOD2023 - International Conference on Machine Learning, Optimization, and Data Science
Side projects
These are some of my favourite public projects. I have other private ones, but you can check these out!
News
She's In The Game Workshop
March 2025I participated as a mentor in the She’s in the Game workshop at Electronic Arts, where we guided fourteen teenage girls through the process of creating their first video game. We introduced them to various aspects of game development and showed them that it can be an exciting and rewarding career path!
Challenge Based Innovation @ CERN
Autumn 2022This project was part of the EU-ATTRACT Academy Program, which aims to explore ways to apply discoveries and breakthroughs from leading European research institutions. We developed a prototype consisting of an electronic device designed to track the precise location of a tumor within a patient’s body, along with software developed in C#. The software reconstructs and visualizes a digital representation of the tumor based on the data received from the device.
BitsxLaMarató
2020, 2021 and 2022 editionsOrganized and directed BitsxLaMarató, a hackathon hosted by FIB, BSC, and Hackers@UPC in support of La Marató de TV3, a public Catalan charity initiative that raises funds for biomedical research projects. Around 150 participants spent a weekend developing technological projects that addressed challenges proposed by institutions such as hospitals and research centers. The event also featured outreach talks and activities aimed at raising awareness and sharing knowledge.
HackUPC
2020 - 2023Organizer of HackUPC, a student hackathon with 700 participants from around the world. I led the Marketing department, managing communications with hackers and running social media.