Machine Learning for Finance & Complex Systems

Spring 2025, ETH Zurich



Info


Content

This course introduces machine learning methods and frameworks that can be used for modelling and analysing complex systems with a particular focus on financial time series. It has two main objectives: (i) theoretical - to provide an overview of machine learning methods with a focus on complex systems and financial time series; (ii) practical - to allow students to gain practical experience by working on a coding project based on a theoretical topic of part (i).

The following topics will be covered: complex systems, empirical facts in finance, PyTorch, ensemble learning, neural networks, clustering, graph cuts, matrix factorisation, reinforcement learning, MCMC, LSTM, attention mechanism, neural ODEs, PINNs, transformers, Black–Litterman model


Books

  1. Paszke, Adam, et al. "Pytorch: An imperative style, high-performance deep learning library." Advances in neural information processing systems 32 (2019).
  2. Ian Goodfellow, Yoshua Bengio and Aaron Courville (2020). Deep Learning. MIT Press.
  3. Mehta, Pankaj, et al. "A high-bias, low-variance introduction to machine learning for physicists." Physics reports 810 (2019): 1-124.
  4. Tsay, Ruey S. Analysis of financial time series. John wiley & sons, 2005.
  5. Richmond, Peter, Jürgen Mimkes, and Stefan Hutzler. Econophysics and physical economics. Oxford University Press, USA, 2013.