Marcus Voss

Marcus Voss

Intelligence Architect and AI Expert, PhD Student

Birds on Mars

TU Berlin (DAI-Lab)

Climate Change AI

CorrelAid

Biography

I am passionate about using machine learning, AI, and data-driven methods to support the energy transition and climate change mitigation and adaptation.

I am an Intelligence Architect and AI Expert at Birds on Mars, where I enable companies to create value with their data.

At the Distributed Artificial Intelligence Laboratory (DAI-Lab), I have coordinated and worked on several research projects investigating how digitization and AI can support the energy transition. Within those projects, I have been modeling, forecasting, and optimizing different demand-side processes such as electric vehicles, smart building- and smart home loads, and renewable generation. I coordinated the research group Smart Energy Systems, where I was responsible for aligning the DAI research for solutions for the energy system. I have co-supervised several Seminar, Bachelor and Master theses and project-based courses, mostly in applied machine learning for energy data, hoping to inspire more students to work in the field.

In my doctoral research, I work on analyzing low voltage-level smart meter data using non-Euclidean distance measures and neural networks with applications in load forecasting and load profile clustering. Are you also analyzing load data? Then check out or contribute to our list of load datasets!

I volunteer in Climate Change AI’s content committee to provide resources for researchers and practitioners transitioning into AI and machine learning for climate change mitigation and adaptation, like on the CCAI Wiki.

Interests
  • Sustainable use cases of AI and Machine Learning
  • Applications of AI and ML in energy systems
Education
  • PhD in Computer Science, 2021 (planned)

    TU Berlin

  • M.Sc. in Information Systems, 2014

    Humboldt University of Berlin

  • B.Sc. in Information Systems, 2011

    HWR Berlin

Recent & Upcoming Talks

Recent Publications

Quickly discover relevant content by filtering publications.
(2022). Short-Term Density Forecasting of Low-Voltage Load using Bernstein-Polynomial Normalizing Flows. arXiv.

PDF Cite

(2021). Nachhaltigkeitskriterien für künstliche Intelligenz - Entwicklung eines Kriterien- und Indikatorensets für die Nachhaltigkeitsbewertung von KI-Systemen entlang des Lebenszyklus. IÖW-Schriftenreihe 220/21.

PDF Cite

(2021). Probabilistic Short-Term Low-Voltage Load Forecasting using Bernstein-Polynomial Normalizing Flows. Workshop Tackling Climate Change with Machine Learning at ICML.

Cite

(2021). DIN SPEC 91410-2:2021-05: Energieflexibilität – Teil 2: Identifizierung und Bewertung von Flexibilität in Gebäuden und Quartieren. Beuth Verlag.

PDF Cite DOI

(2020). Integration of Building Inertia Thermal Energy Storage into Smart Grid Control. In SEST 2020.

PDF Cite DOI

Contact