skip to content

Department of Computer Science and Technology

  • PhD Student

I am a second-year PhD student and Gates Scholar in Computer Science (Energy and Environment Group). I am supervised by Srinivasan Keshav. My research investigates the use of Machine Learning for the decarbonisation of buildings. I am particularly interested in the use of in-context reinforcement learning for the decarbonisation of building operations.

Previously, I have explored the design and operation of sustainable energy systems for residential homes, incorporating technologies such as solar PV, stationary batteries, and bidirectional EVs. My work led to the development of SOPEVS, an innovative algorithm that personalises the joint sizing and operation of solar PV and storage for single-family homes. Additionally, I created SPAGHETTI, a synthetic data generation tool for modelling EV usage traces.

Currently, my research is centred on combining reinforcement learning with decision transformers to optimise the operation of HVAC systems, energy storage and EV charging in commercial buildings. 

Check out my website (https://amcberkes.github.io/) for a short overview of my research.

Biography

Career

2021-2022: Product Team, Archlet.  

2021-2022: Partner, Wingman Campusfund.

2021-2022: Teaching assistant. ETH Zürich Department of Computer Science.

2019: Machine Learning software development intern. Ghelia - Sony Computer Science Laboratories.

Qualifications

2023: Bsc. in Computer Science, ETH Zürich

Awards

Gates Cambridge Scholar

Heidelberg Laureate Forum 

Lauréate du Concours Général

Research

Building control and operation, Reinforcement Learning, decision transformers, in-context Reinforcement Learning

Publications

  • Berkes, Anaïs. "HVAC-DPT: A Decision Pretrained Transformer for HVAC Control."38th Conference on Neural Information Processing Systems (NeurIPS 2024) - Tackling Climate Change with Machine Learning Workshop & Women in Machine Learning (WiML) Workshop  (2024).
  • Anaïs Berkes and Srinivasan Keshav. 2024. SOPEVS: Sizing and Operation of PV-EV-Integrated Modern Homes. In Proceedings of the 15th ACM International Conference on Future and Sustainable Energy Systems (e-Energy '24). Association for Computing Machinery, New York, NY, USA, 14–26. https://doi.org/10.1145/3632775.3661941
  • Berkes, A. and Keshav, S., 2024. SPAGHETTI: a synthetic data generator for post-Covid electric vehicle usage. _Energy Informatics_, _7_(1), pp.1-21.

Contact Details

Room: 
FW01
Email: 

amcb6@cam.ac.uk