Additional Chapters of Machine Learning
Start: February 8, 2022
Learning Objectives & Outcomes
Program of the course
Studying the basic algorithms in data analysis, forecasting and machine learning necessary for reading literature and building your own intelligent systems in the financial field.
- Introduction to machine learning.
- Dimension reduction.
- Anomaly detection. Unbalanced classification.
- Multi-armed bandits and RL.
- Probabilistic approach to machine learning. Bayesian linear regression.
- Nuclear methods. Regression based on Gaussian processes. RKHS space.
- Optimization and active learning based on surrogate models.
- Neural networks. Estimation of parameters of deep neural networks.
- Convolutional neural networks. The use of such models in practice.
- Presentation training. Using self-learning to get insights.
- Recurrent neural networks. The mechanism of attention and transformers.
- Building ensembles of machine and deep learning models.
- Modern generative models. GANs, optimal transport.
- Classical models for working with time series: ARIMA and decomposition of the series into components.
- The use of machine learning in the financial field.