Alejandro Lancho

Associate Professor · Signal Theory and Communications
Universidad Carlos III de Madrid
Signal Processing Group · Leganés, Madrid

Email Google Scholar ORCID arXiv GitHub LinkedIn YouTube

Before returning to Universidad Carlos III de Madrid I was a postdoctoral researcher at Chalmers University of Technology, and then a Marie Skłodowska-Curie Global Fellow at MIT.

I teach Deep Learning, Information Theory for Machine Learning, and Natural Language Processing at UC3M.

I am a member of the IEEE Information Theory Society, where I serve as Vice President of its Spain Chapter and sit on the Student and Outreach Committee. Through the committee I co-organize international data-science competitions, most recently at ESIT 2026 and ISIT 2026. Separately, as a postdoc at MIT, I co-organized the ICASSP 2024 Grand Challenge on data-driven radio-frequency signal separation.

First Prize, IEEE ISIT Data-to-Information Challenge — 2025, with team ISAC Innovators
Best Student Paper Award — IEEE MLSP 2022
Third Prize, Student Paper Contest — Asilomar 2022
Finalist, Jack Keil Wolf Student Paper Award — IEEE ISIT 2017

Research

Finite-blocklength information theory

Limits of reliable communication when only a few channel uses are available.

Deep learning for communications

Signal separation, interference rejection, detection, and distributed inference for tasks such as localization.

Machine learning, fundamentals to applications

Method development, and its use in applied domains including health.

Selected work

All publications →

2026 A Probabilistic Hard Concept Bottleneck for Steerable Generative Models ICLR OpenReview — A Probabilistic Hard Concept Bottleneck for Steerable Generative Models
2025 RF Challenge: The Data-Driven Radio Frequency Signal Separation Challenge IEEE OJ-COMS arXiv — RF Challenge: The Data-Driven Radio Frequency Signal Separation Challenge IEEE Xplore — RF Challenge: The Data-Driven Radio Frequency Signal Separation Challenge
2023 Score-based Source Separation with Applications to Digital Communication Signals NeurIPS arXiv — Score-based Source Separation with Applications to Digital Communication Signals OpenReview — Score-based Source Separation with Applications to Digital Communication Signals Code on GitHub — Score-based Source Separation with Applications to Digital Communication Signals
2021 URLLC with Massive MIMO: Analysis and Design at Finite Blocklength IEEE Trans. Wireless Commun. arXiv — URLLC with Massive MIMO: Analysis and Design at Finite Blocklength IEEE Xplore — URLLC with Massive MIMO: Analysis and Design at Finite Blocklength