Robert Bayer

Robert Bayer

Ph.D. Student

IT University of Copenhagen

Hello!

My name is Robert Bayer and I am a Ph.D. student in the Resource-Aware Data Systems (RAD) group at the IT University of Copenhagen (ITU), Denmark. My research focuses on unlocking power of machine learning on small, resource-constrained hardware. Specifically I focus on challenges in resource management, task colocation and performance analysis.

Before starting my Ph.D. I worked at the ITU as a research assistant, while working towards my master's degree in computer science, which I finished in 2023. Before starting my master's, I got my bachelor's degree in data science from the same university in 2021. Throughout my studies, including currently during Ph.D., I was fortunate to be supervised by prof. Pınar Tözün.

Interests
  • TinyML
  • Space IoT
  • Edge computing
Education

  • Ph.D. in Computer Science, 2023 - present

    IT University of Copenhagen

  • MSc. in Computer Science, 2021 - 2023

    IT University of Copenhagen

  • BSc. in Data Science, 2018 - 2021

    IT University of Copenhagen

Publications

Robert Bayer, Julian Priest, Pınar Tözün (2024). Reaching the Edge of the Edge: Image Analysis in Space. In DEEM.

PDF Cite Code

Robert Bayer, Jon Voigt Tøttrup, Pınar Tözün (2022). TPCx-AI on NVIDIA Jetsons. In TPC-TC.

PDF Cite Code

Talks

No Resource Left Behind: Machine Learning at the Edge
Talk about past, present and future research conducted during Ph.D, specifically we characterize the hardware landscape at the edge with a special focus on machine learning workloads and use the findings to guide the design of a smart resource manager for edge machine learning systems. Furthermore, we identify the common inefficiencies in these workloads and develop optimization techniques to combat them.
Apr 22, 2024 — Athens, Greece
Image Analysis in Space + Hardware Efficiency
Talk about my master's thesis and the future directions of my research in resource-constrained and -aware machine learning within a context of climate change and climate research.
Dec 11, 2023 — IT University of Copenhagen, Copenhagen, Denmark
Image Analysis in Space
Talk on the past experiences with image analysis on board of small satellites, and possible integration of this topic into Danish higher education.
Sep 29, 2023 — AI Pioneer Center, Copenhagen, Denmark

Experience

 
 
 
 
 
Ph.D Fellow
IT University of Copenhagen
August 2023 – Present Copenhagen, Denmark

Research topics include:

  • TinyML
  • Resource management
  • Task colocation
  • Hardware accelerators
  • Performance analysis
 
 
 
 
 
Research Assistant
IT University of Copenhagen
January 2022 – July 2023 Copenhagen, Denmark
 
 
 
 
 
Senior Software Engineer
GlycoSpot
January 2020 – December 2021 Copenhagen, Denmark
Developed a custom portable spectrophotometer, used for in-field enzyme activity screening. This method resulted in two patents.
 
 
 
 
 
Software Engineer
GlycoSpot
October 2017 – December 2019 Copenhagen, Denmark
 
 
 
 
 
Full-stack Web Developer
.upvision
November 2016 – June 2017 Senica, Slovakia

Teaching & Supervision

Teaching

Internet of Things (Spring 2024)

  • Embedded systems and programming
  • Data analytics: ML basics
  • Edge computing & TinyML

CubeSats 101 (Summer 2023)

  • CubeSat hardware architecture
  • CubeSat programming

Supervision

Master's theses

  • DIPP - A DISCO-2 Image Processing Pipeline, Spring 2024

    Students: Nikolaj Sørensen, Jeppe Lindhard and Daniel Kjellberg

    Co-supervised with Pınar Tözün

  • Remotely Programmable Heterogeneous Multi-Processsing Software System Setup in CubeSats, Spring 2024

    Students: Nicolaj Valsted

    Co-supervised with Pınar Tözün

Research projects

  • Empowering Earth Observation Research: A Modular Image Processing Pipeline for CubeSats, Fall 2023

    Students: Nikolaj Sørensen, Jeppe Lindhard and Daniel Kjellberg

    Co-supervised with Julian Priest

  • Simulation of CubeSat Task Scheduling Strategies for Earth Observation, Fall 2023

    Students: Nicolaj Valsted

    Co-supervised with Julian Priest

Bachelor's theses

  • Understanding and Evaluating Data Drift in Computer Vision from Edge Computing Perspective, Spring 2024

    Student: Janusz Jakub Wilczek Co-supervised with Pınar Tözün

Accomplish­ments

Best Computer Science Master's Thesis in Denmark in 2023
IDA and Dansk Industri Nov 2023
I have won the annual award for the best computer science master’s thesis in Denmark for my work on evaluation, design and implementation of machine learning-based image processing pipeline deployed on board of the DISCO1 cubesat.