Milos Tomic

Milos Tomic

Machine Learning Expert

Biography

Milos Tomic studied Mathematics & Statistics at University of Warwick where he focused on mathematical analysis (PDE’s), probability theory, stochastic analysis, and Bayesian statistics. He currently works at a Quantitative Hedge Fund as a Quantitative Researcher (Quant). At his work he researches and builds sophisticated trading strategies by implementing stochastic and machine learning models (ML) for statistical arbitrage. Previously he was a Data Scientist building ML models from massive datasets that were used to increase the volume of mortgage originations.

His main research areas are:

  • Machine Learning - especially kernel methods and generalisations to infinite-dimensional spaces, and neural networks.
  • Stochastic Analysis
  • Partial Differential Equations (PDE’s)
  • Dimensionality Reduction
  • Hawkes Processes

He has a strong experience in:

  • Python
  • R Language
  • C++

Projects Worked On

  • Statistical arbitrage of financial instruments using stochastic methods and mean-reversion models.
  • Modelling volatility via Computational Stochastic Differential Equations.
  • Clustering financial time-series based on risk.
  • Built a custom unsupervised ML model that was trained by massive datasets, such as UK’s mortgage search engine results. This resulted in optimised mortgage rates that increased company’s volume of originations.
  • Research includes generalising the Tikhonov regularisation to infinite-dimensional spaces and thus enabling its use in kernel-based methods in Machine Learning.
  • Worked on cutting-edge pattern recognition in data.