Nicola Pezzotti

Software Engineer and Researcher in Visual Data Mining and Machine Learning

Follow me on GitHub

About me

I’m a PhD candidate in the Computer Graphics & Visualization group at the Delft University of Technology. I’m supervised by Prof. Anna Vilanova, my Promotor (doctoral advisor) is Prof. Elmar Eisemann and Prof. Boudewijn P.F. Lelieveldt is my co-Promotor. My research interest is mainly oriented towards the application of Machine Learning algorithms, in particular Manifold Learning and Deep Learning, in a Visual Analytics context. I’m also interest in 3D Scanning Technologies and Geometry Processing.

I’m familiar with different programming languages and technologies. I generally work in C++/CUDA for performance reasons, however I’m also familiar with [Open|Web]GL, Python, TypeScript, JavaScript and Matlab. I strongly believe that high-quality code is of main importance also in academia.

Projects

High-Dimensional Inspector

C++ library for scalable and interactive high-dimensional data analysis. The core feature of HDI is the implementation of the Approximated-tSNE and Hierarchical-SNE, which together allow the visual analysis of millions of high-dimensional data points on a desktop computer. The library also includes several visualizations, analytical workflows and data analysis algorithms.

TensorFlow.js tSNE

tfjs-tsne is a module of the TensorFlow.js library that implements the tSNE algorithm. tfjs-tsne makes use of a WebGL trick to accelerate the gradient computation and the can be run in the client side of the web browser. Check out the demo on the MNIST dataset. This project featured on the Google AI Blog.

Cytosplore

Cytosplore is an interactive visual analysis system for understanding how the immune system works. The goal of the analysis framework is to provide a clear picture of the immune systems cellular composition and the cells’ corresponding properties and functionality.

Selected Publications

A detailed list of my publications can be found at this page

  • N. Pezzotti, A. Mordvintsev, T. Höllt, B.P.F. Lelieveldt, E. Eisemann, A. Vilanova. Linear tSNE Optimization for the Web. ArXiV preprint, 2018. PDF, Demo, Google AI post
  • V. van Unen*, T. Hollt*, N. Pezzotti*, N. Li, M. J.T. Reinders, E. Eisemann, A. Vilanova, F. Koning, B. P.F. Lelieveldt. Interactive Visual Analysis of Mass Cytometry Data by Hierarchical Stochastic Neighbor Embedding Reveals Rare Cell Types. Nature Communications, 2017. Website PDF
  • N. Pezzotti, T. Höllt, J. van Gemert, B.P.F. Lelieveldt, E. Eisemann, A. Vilanova. DeepEyes: Progressive Visual Analytics for Designing Deep Neural Networks. Transaction on Visualization and Computer Graphics (Proceedings of IEEE VIS 2017), 2018. PDF, Video
  • N. Pezzotti, T. Höllt, B.P.F. Lelieveldt, E. Eisemann, A. Vilanova. Hierarchical Stochastic Neighbor Embedding. Computer Graphics Forum (Proc. of EuroVis), 2016. PDF, Suppl. Mat., Video, Slides
  • T. Höllt, N. Pezzotti, V. van Unen, F. Koning, E. Eisemann, B.P.F. Lelieveldt, A. Vilanova. Cytosplore: Interactive Immune Cell Phenotyping for Large Single-Cell Datasets. Computer Graphics Forum (Proc. of EuroVis), 2016. PDF
  • N. Pezzotti, B.P.F. Lelieveldt, L. van der Maaten, T. Höllt, E. Eisemann, and A. Vilanova. Approximated and User Steerable tSNE for Progressive Visual Analytics. Transaction on Visualization and Computer Graphics (Presented at IEEE VIS 2016), 2016. PDF, Suppl. Mat., Video I, Video II, Video III

Awards

  • 2006 - Awarded in “Expo Del capitale umano 2006” for academic merit
  • 2005 - Silver medal in the Italian selection for the International Olympiad in Informatics (IOI)