Nahuel Costa, PhD

Nahuel Costa, PhD

Machine Learning researcher 🤖 Assistant professor

University of Oviedo

Biography

My research focuses on maximizing the potential of Machine Learning, particularly in scenarios with limited data, to anticipate potential outcomes and support decision-making in monitoring systems. This includes applications across both biomedical and industrial domains. In the biomedical field, I have worked on problems such as arrhythmia detection and classification, as well as the analysis of X-ray images and functional PET images from patients in deep coma. In the industrial domain, my work has included monitoring and prognostics for lithium-ion batteries, aircraft engines, and industrial fans.

I have a particular interest in developing ML models that are robust and effective but also interpretable and accessible to non-experts, ensuring their practical utility across diverse applications. As an educator, I aim to convey my enthusiasm to students and provide them with the necessary tools to define and pursue their own goals.

💬 Feel free to reach out to me if you are interested in my research, looking for colaboration, or just for some interesting discussion.
✉️ You can shoot me a message at costanahuel@uniovi.es or any of my other social networks, I’ll try to respond as soon as I can!

Interests
  • Prognosis & Health Management
  • Generative models (GANs, VAEs, Flows, Transformers, Difussion)
  • Domain adaptation
  • Survival prediction
  • Explainable AI
  • Conformal prediction
  • LLMs and RAG
Education
  • PhD in Artificial Intelligence, 2023

    University of Oviedo

  • MSc in Computer Science, 2020

    University of Oviedo

  • BSc in Computer Science, 2019

    University of Oviedo

Research Experience

 
 
 
 
 
University of Oviedo
Assistant Professor
University of Oviedo
Sep 2023 – Present

Supervising several bachelor’s theses and co-supervising a Ph.D. thesis in “Prognosis of Degenerative Diseases Using Unsupervised and Partially Supervised Learning Techniques”

Subjects I teach:

Apuntes de redes generativas 🇪🇸

Apuntes de minería de texto 🇪🇸

Apuntes de algoritmia 🇪🇸

Innovation Projects:

  • Automated Moodle quiz creation and optimization using learning analytics, application to learning programming languages

 
 
 
 
 
University of Oviedo
Lecturer
University of Oviedo
Feb 2021 – Aug 2023

Co-Supervisor for four BA thesis (two of them obtained with highest honors)

Subject I taught:

  • Business Intelligence
  • Data Visualization
  • Algorithmics
  • Operating Systems
  • Databases
  • Programming methodology
  • Introduction to programming
 
 
 
 
 
University of Montpellier
Visiting Researcher
University of Montpellier
May 2023 – Jul 2023 Montpellier
Research and development of Machine Learning diagnostic methods for positron emission tomography (PET) images at the Laboratoire d’informatique, de robotique et de microélectronique de Montpellier (LIRMM).
 
 
 
 
 
University of Hawaii at Manoa
Visiting Researcher
University of Hawaii at Manoa
May 2022 – Sep 2022 Hawaii
Research and development of Machine Learning tools for the diagnosis and prognosis of lithium-ion batteries at Hawaii Natural Energy Institute (HNEI).
 
 
 
 
 
University of Oviedo
Research technician
University of Oviedo
Oct 2019 – Jan 2021
Development of computational health models for the treatment of rechargeable batteries.
 
 
 
 
 
University of Oviedo
Research intern
University of Oviedo
Oct 2019 – Jan 2021
Analysis of intracardiac electrocardiograms for the prediction of cardiovascular diseases.

Tech stack


python, matlab, R, C, C++, Java

TensorFlow, PyTorch, Keras, HF Transformers, HF Diffusers, LangChain, LlamaIndex

LaTeX

Partnerships & Collaborations

Total Energies Logo

Zitrón Logo

Huca Logo

Consejería Salud Logo

Medtronic Logo

Izertis Logo

Publications

(2025). Integrating imprecise data in generative models using interval-valued Variational Autoencoders. In Information Fusion.

PDF Cite Code Project DOI

(2024). Few-shot generative compression approach for system health monitoring. In Reliability Engineering & System Safety.

PDF Cite Code Project DOI

(2024). Rapidae: A Python library for creation, experimentation, and benchmarking of Autoencoder models. In IEEE WCCI.

PDF Cite Code DOI

(2023). ICFormer: A Deep Learning model for informed lithium-ion battery diagnosis and early knee detection. In Journal of Power Sources.

PDF Cite Code Dataset Project DOI

(2023). Learning remaining useful life with incomplete health information: A case study on battery deterioration assessment. In Array.

PDF Cite Project DOI

(2023). Physics-informed learning under epistemic uncertainty with an application to system health modeling. In International Journal of Approximate Reasoning.

PDF Cite Project DOI

(2023). Simplified models of remaining useful life based on stochastic orderings. In Reliability Engineering & System Safety.

PDF Cite Project DOI

(2023). Enhancing Time Series Anomaly Detection Using Discretization and Word Embeddings. In SOCO.

PDF Cite Project DOI

(2023). Data-Driven Diagnosis of PV-Connected Batteries: Analysis of Two Years of Observed Irradiance.

PDF Cite Project DOI

(2023). Data-driven direct diagnosis of Li-ion batteries connected to photovoltaics.

PDF Cite Code Project DOI

(2022). Li-ion battery degradation modes diagnosis via Convolutional Neural Networks. In Journal of Energy Storage.

PDF Cite Code Dataset Project DOI Demo

(2022). Weakly Supervised Learning of the Motion Resistance of a Locomotive Powered by Liquefied Natural Gas. In SOCO.

PDF Cite Project DOI

(2022). Informed Weak Supervision for Battery Deterioration Level Labeling. In IPMU.

PDF Cite Project DOI

(2022). Variational encoding approach for interpretable assessment of remaining useful life estimation. In Reliability Engineering & System Safety.

PDF Cite Code Dataset Project DOI Demo

(2022). RUL-RVE: Interpretable assessment of Remaining Useful Life. In Software Impacts.

PDF Cite Code Dataset Project DOI Demo

(2020). Graphical analysis of the progression of atrial arrhythmia using recurrent neural networks. In International Journal of Computational Intelligence Systems.

PDF Cite Code Project DOI

Contact