Avatar

Daniele Gammelli

PhD Student

Machine Learning for Smart Mobility, DTU

About me

I’m a PhD student in Machine Learning and Operations Research at the Technical University of Denmark (DTU). I work with Francisco C. Pereira, Dario Pacino and Filipe Rodrigues on applying pattern recognition, machine learning and optimization to understand and predict mobility behaviour within the Machine Learning for Smart Mobility group.

Before joining DTU, I graduated from the University of Rome “La Sapienza” with a MSc in Quantitative Methods for Decision Making within the Management Engineering Degree. Back in Rome I worked on my MSc thesis by applying machine learning and probabilistic models to health applications (in particular we worked on “Multiple Sclerosis course prediction”).

I also worked in the Operations Research Team of Amazon EU within Amazon Transportation Services where we applied Optimization and Machine Learning models for the planning of EU’s truck schedules.

I am excited about improving science towards a more human-level AI through machine learning. Feel free to get in touch for any intersting discussion ;-)

Interests

  • Machine Learning and Pattern Recognition
  • Probabilistic Graphical Models and Bayesian inference
  • Deep Learning
  • Generative Models
  • Reinforcement Learning

Education

  • PhD in Machine Learning and Operations Research, 2021

    Technical University of Denmark

  • MSc in Management Engineering (Quantitative Methods for Decision Making), 2018

    University of Rome "La Sapienza"

  • BSc in Management Engineering, 2016

    University of Rome "La Sapienza"

Experience

 
 
 
 
 

PhD Student in ML and OR

Technical University of Denmark

Jan 2019 – Present Denmark
 
 
 
 
 

Operations Research Intern

Amazon

Aug 2018 – Jan 2019 Luxembourg
 
 
 
 
 

MSc Degree

University of Rome ‘La Sapienza’

Sep 2016 – Jul 2018 Rome

Publications

Estimating Latent Demand of Shared Mobility through Censored Gaussian Processes

Transport demand is highly dependent on supply, especially for shared transport services where availability is often limited. As …

A Machine Learning Approach to Censored Bike-Sharing Demand Modeling

Transportation demand is highly dependent on supply, especially for shared transport modes where the availability of the service is …

Blog Posts

Recent & Upcoming Talks

Lectern Session @TRB2020 - Gaussian Processes for Censored Demand Estimation

Lectern Session on Travel Modeling Techniques

Contact

  • daga@dtu.dk
  • DTU Building 116, Kgs. Lyngby, 2800, Denmark