Funded master thesis in Probabilistic Machine Learning group
The Probabilistic Machine Learning group (PML) at Aalto University and Elisa Corporation are looking for a master student to work on a challenging machine learning problem.
The position is fully funded by Elisa and it is filled as soon as a suitable candidate is found. It is also possible for students who are not yet in their MSc thesis stage to apply.
If you are Interested, please send an email to the email addresses below and attach your CV, study records, and a brief justification for why you are interested. In the application please elaborate past experience in programming and machine learning.
The ideal student should be able to contribute both on the theory and application side of the project. This includes design and implementation of new algorithms, working with real data, and writing papers, which are necessary experience for both future doctoral studies and/or an industry career in data science.
Title: Multi-armed-bandit utilization on marketing automation
Abstract: The goal of multi-armed bandit problem1 is to find the best decision, among a set of all possible decisions with unknown outcomes, with minimum number of trials and errors. The solution algorithm randomly explores the decision space to gain knowledge on how good the different decisions are while simultaneously exploiting more promising decisions based on the learned knowledge. In this project, the student designs, tests, and implements a specific multi-armed bandit algorithm that meets the requirements of the Elisa’s marketing automation platform or in the other case prepare the way how Elisa can implement the algorithms based on adaptive prediction modelling and Bayesian modelling.
Keywords: Multi-armed bandit problem; exploration and exploitation; intent modeling; probabilistic user models
Contact persons: PhD Neda B. Marvasti (neda.marvasti [at] aalto [dot] fi), MSc Pedram Daee (pedram.daee [at] aalto [dot] fi)
Supervisors: Prof. Samuel Kaski (Aalto), Juha Lehtiranta (Elisa)