Ekaterina Abramova
Adjunct Assistant Professor of Management Science and Operations
PhD (Imperial College London) MSc (Birkbeck, University of London) BSc (Loughborough University)
Ekaterina Abramova holds a PhD in Artificial Intelligence and Machine Learning from Imperial College London, an MSc in Financial Engineering from Birkbeck University of London, and a BSc in Chemistry with Forensic Analysis from Loughborough University. Her latest research focuses on the discovery of trading strategies in electricity markets. Ekaterina’s research during her PhD focused on the Reinforcement Learning branch of Machine Learning, with a particular topic of solving nonlinear optimal control problems. Her work has been included in various journals including Multidisciplinary Digital Publishing Institute (MDPI) Energies, European Workshop on Reinforcement Learning (EWRL) and Imperial College Computing Student Workshop (ICCSW). Ekaterina is interested in discovering statistical arbitrage with the help of econometric and machine learning techniques as applied to financial data analysis and forecasting. Her personal interests are in the space of innovation and technology, with particular focus on the blockchain, decentralisation and the transformation of data ownership offered by web3.0.
- eabramova@london.edu
- +44 (0)20 7000 8835
PhD Awards
- Cognitive Robotics Conference: Best Presentation Award 2014
- Google: PhD Poster Competition Award 2014
- Imperial College Graduate Research Symposium, Highly Commended Poster 2014
- Cosyne Conference: Qualcomm Travel Award ($500) 2013
BSc Awards
- Departmental Prize for Best Academic Performance 2003
2021
Optimal Daily Trading of Battery Operations using Arbitrage Spreads
Abramova E; Bunn D W
Energies 2021 Vol 14:16 p e4931
2020
Forecasting the intra-day spread densities of electricity prices
Abramova E; Bunn D
Energies 2020 Vol 13:3 p 687
2012
Hierarchical, Heterogeneous Control of Non-Linear Dynamical Systems using Reinforcement Learning
Abramova E; Dickens L; Kuhn D; Faisal A
Journal of Machine Learning Research 2012 Vol 1-10
2011
Combining Markov Decision Processes with Linear Optimal Controllers
Abramova E; Kuhn D; Faisal A
Proceedings of ICCSW ‘11 September 29th–30th 2011, London UK
2019
RLOC: Neurobiologically inspired hierarchical reinforcement learning algorithm for continuous control of nonlinear dynamical systems
Abramova E; Dickens L; Kuhn D; Faisal A
ARXIV
2016
Combining Reinforcement Learning and Optimal Control for the Control of Nonlinear Dynamical Systems
Abramova E
(PhD Thesis)
Teaching portfolio
Our teaching offering is updated annually. Faculty and programme material are subject to change.
-
Masters Degrees core courses
A key part of our Masters programmes curriculum.
View courses Show lessShow less-
Data Analytics for Management
Data Analytics for Management -
Financial Reporting Analytics
Financial Reporting Analytics PROGRAMMES WITH THIS CORE COURSE
-
Introduction to Python for Data Science
Introduction to Python for Data Science PROGRAMMES WITH THIS CORE COURSE
-
Machine Learning for Big Data
Machine Learning for Big Data PROGRAMMES WITH THIS CORE COURSE
-
Data Analytics for Management
-
Masters Degrees electives
Optional courses providing a deep dive into specialist areas.
View electives Show lessShow less-
No Code Data Science (with Tableau and Alteryx)
In this course students delve deep into the world of data analytics without writing a single line of code. Through compelling case studies and real-world datasets, participants will master advanced visualization techniques and explore a spectrum of machine learning methodologies, from supervised and unsupervised learning to text classification and predictive analytics. With hands-on sessions using powerful tools like Tableau and Alteryx, students will emerge with a robust analytical toolkit, ready to tackle large datasets with confidence. PROGRAMMES WITH THIS ELECTIVE
-
Python for Finance
Students will learn Python coding and will apply the skills in a variety of financial problems, such as devising and evaluating trading strategies, portfolio construction, and risk assessment, among others. Students are expected to have basic level of coding experience in any other language and a good technical background (namely should be familiar with linear regression and basic linear algebra). PROGRAMMES WITH THIS ELECTIVE
-
Using Analytics for Social and Environmental Impact
The primary objective of this course is to enhance the participants’ ability to use analytics to develop insights in a wide range of business situations. We will work on five in-depth real-world case studies developed by the instructor. For each case, we will start with breaking down what seems to be a complex and ill-structured problem to simpler subproblems that are amenable to analysis. Through a mixture of lectures and workshops, we will use data analysis (visualization, regression) and modelling techniques (Monte-Carlo simulation, decision-tree analysis, optimization) to build progressively more sophisticated models that allow us to evaluate possible alternatives, make sensible recommendations, and serve as the basis for communicating findings and influencing decisions. After each session, participants will have the opportunity to further refine the models developed in class and prepare a presentation for the management of the company featured in the case study. PROGRAMMES WITH THIS ELECTIVE
-
No Code Data Science (with Tableau and Alteryx)
-
Show less
Executive Education
Short programmes offering academic excellence, global focus and exceptional diversity of perspective.
View programmes Show less