Simão Lopes Eduardo

Hi! I’m currently looking for jobs in industry!

I’m a Machine Learning PhD Graduate (class of 2023) from the University of Edinburgh, School of Informatics and Institute for Adaptive and Neural Computation (IANC). My main supervisor was Dr. Charles Sutton, and my secondary was Prof. Chris Williams. My research interests were on Generative Models and Deep Learning under Data Corruption, and generally Robust Machine Learning.

My PhD thesis – Data Cleaning with Variational Autoencoders – was on jointly automating Outlier Detection and Data Repair tasks on corrupt datasets using Variational Autoencoders, a type of deep generative model. In a ML pipeline, these two tasks are commonly needed together in the Data Cleaning step. See thesis here.

More recently, I was a ML Researcher at Hazy (UK startup), a company using generative models to produce synthetic data that complies with differential privacy.

Scroll down for publications.

More about me…

I have a BSc. and MSc. in Electrical and Computer Engineering from Instituto Superior Tecnico (IST) in Portugal. My MSc thesis was on radio resource management in wireless networks, where I learnt about constrained convex optimization, integer programming, queuing theory.

I have a MSc(R) in Data Science from University of Edinburgh, specializing in Machine Learning. My thesis focused on combining probabilistic modelling and logic rules (integrity constraints) for outlier detection.

I was a GNSS Software Engineer at GMV (Lisbon) working on digital signal processing algorithms for GNSS receivers (e.g. GPS, Galileo). I worked on time-series data and signal estimation / forecasting models.

I was a Research Assistant at IST in convex optimization applied to image processing, using proximal algorithms.