About me
Hi, I'm Samuel. I am a master's student in Physics and Data Science at EPFL (École polytechnique fédérale de Lausanne, if you like the sound of French). I live in Switzerland, next to the Geneva lake, with a beautiful view on the Alps. My main areas of interest are Machine Learning and Quantum Computing
I am originally from Germany but spent almost all of my life in other countries (Switzerland being my 9th country). When I was 5 years old, my family moved to Croatia. There I finished elementary and high school. I also went to school in South Africa and England for a while, as an exchange student. After graduating from high school, I worked as a volunteer teacher in an orphanage in Kenya, near Nairobi. During my undergraduate studies, I worked in Ice and Climate research at the University of Copenhagen in Denmark (see publications). Afterwards, I joined the hedge fund D. E. Shaw & Co. in London and New York City for a while, where I created computer simulations of the bond and foreign exchange market.
Before starting my master's degree at EPFL, I spent some time in the sunny and picturesque city of La Jolla, at University of California in San Diego. There, I worked on Machine Learning research (see publications) and enjoyed surfing in the Pacific Ocean.
Since September 2018, I am studying Physics and Data Science at EPFL in Switzerland. As a research assistant, I work in the group of Prof. Giovanni De Micheli on Quantum Computing. I am founder and president of the EPFL Quantum Computing Association and vice-president of the Debate and Model United Nations association. At the moment, I am working at the National University of Singapore on Quantum Optimization Algorithms, with applications in finance.
CV
EDUCATION
From 2018
EPFL (École polytechnique fédérale de Lausanne), Switzerland
- Master of Science: Physics & Data Science
- EPFL Excellence Fellowship (top 1%)
- Specialization: Quantum Computing & Machine Learning
- Associations and Societies:
-
- Founder and President of the Quantum Computing Association
-
- Vice-President of the Debating & Model United Nations Association (MUN)
- Member of the German Academic Scholarship Foundation (Studienstiftung)
2014 - 2017
University of Zagreb, Croatia
- Undergraduate studies: Physics
- Final GPA: 4.9/5.0
- Elected as Student Representative in 2017
- Associations and Societies:
-
- Founder of the Debate Society (politics, philosophy, economy, ...)
- Member of the German Academic Scholarship Foundation (Studienstiftung)
- Project:
-
- Simulation of a Tokamak fusion reactor
2010 - 2014
XV. Gymnasium (MIOC), Zagreb, Croatia
- Focus on Physics, Mathematics and Computer Science
- International Physics Olympiads (IPhO): Kazakhstan 2014, Denmark 2013 and Estonia 2012: 3x Bronze Medal
- Highest scores at the final exams in the Republic of Croatia (100%)
- SAT Subject Tests Math (100%) and Physics (100%)
WORK EXPERIENCE
Summer 2019
National University of Singapore (Visiting Researcher)
- Research in Quantum Optimization Algorithms
- Work at the Centre for Quantum Technologies in the group of Dr. Patrick Rebentrost and Prof. Miklos Santha
Since 2019
EPFL, Lausanne, Switzerland (Research Assistant)
- Implementation of optimization and logic synthesis algorithms for Qiskit-Terra (an open source Python library for Quantum Computers maintained by IBM)
- Work in the group of Prof. Giovanni de Micheli and Dr. Mathias Soeken
Since 2018
University of California San Diego, USA (Visiting Researcher)
- 2 publications (ICCAD 2019, ISLPED 2019) & 4 further submitted papers
- Machine learning research: Development of fast and energy efficient algorithms for specialized computer architecture
- Optimization of algorithms for Natural Language and Human Activity Recognition
2017 - 2018
EPFL, Lausanne, Switzerland + Technical University Vienna, Austria (Internship)
- Solid State physics research: discovering and analyzing new materials
- Work on Quantum Cryptography algorithms
2016
D. E. Shaw & Co., London, UK & New York City, USA
- Monte Carlo simulations of complex derivatives in Python
- Modeling changes in the foreign exchange and bond market
2015
Niels Bohr Institute, University of Copenhagen, Denmark (Research Internship)
- 1 publication (Climate of the past) & 1 further publication in progress
- Data analysis/data mining in Climate science
2014 - 2017
University of Zagreb, Croatia (Student assistant, Physics competitions)
- Organization of the preparations for the International Physics Olympiad
- 2 of my students won bronze medals at the International Physics Olympiad and received full scholarships for their studies at University of Cambridge
- Teaching Physics I & II: Classical Mechanics, Electromagnetism (~80 students)
2014
Mully Children’s Family Orphanage, Kenya (Volunteer work)
- Taught computer programming, mathematics and physics to 125 high school students
- Organized programming courses for students, to give them a chance to see what programming is and motivate them to pursue a career in it, as it requires few resources
Until 2015
Music performance and recording of CDs
- Participated as a singer on 3 CD recordings for humanitarian purposes
- Participated in 15 humanitarian musicals and concerts as an actor, singer or technician
COMPETITIONS & AWARDS
Since 2018
EPFL Excellence fellowship
- Awarded to the best master’s students of every department (< top 1%)
Since 2017
Studienstiftung des deutschen Volkes (German Academic Scholarship Foundation)
- Awarded to the best ~0.5% students in/from Germany
- The selection is based on excellent academic records, social engagements, rhetorical and social skills, academic and sport competitions, and other personal accomplishments.
2014
Award for the best high school graduate in Croatia in 2014 (out of ~40 000 students)
- Selection was based on final exam results, science and sport competitions and social commitment
2014
Bronze medal at the 45th International Physics Olympiad 2014 (Kazakhstan)
2014
1st place at the Croatian National Physics Competition
2013
Bronze medal at the 44th International Physics Olympiad 2013 (Denmark)
2013
1st place at the Croatian Championship in Triathlon
2012 - 2016
City of Zagreb scholarship
- Awarded to top ≈1% students in Zagreb (the capital of Croatia)
2012
Bronze medal at the 45th International Physics Olympiad 2012 (Estonia)
COMPUTER SKILLS
- Good in Python and Matlab
- Ok with C/C++, Linux, Pascal, scikit-learn, TensorFlow, Git, Qiskit,...
LANGUAGES
English
Full professional proficiency
German
Bilingual proficiency
Croatian
Bilingual proficiency
French
Elementary proficiency
SELECTED MEDIA APPEARANCES
The following articles are in Croatian, so you might need to use Google Translate to understand them
Dnevnik nove TV (the most watched news program in Croatia)
Dobro jutro, Hrvatska (the most watched morning program in Croatia)
PUBLICATIONS
ACCEPTED PEER REVIEWED PUBLICATIONS
2019
QuantHD: A Quantization Framework for FPGA Acceleration of Hyperdimensional Computing
- Journal: TCAD - IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Work at University of California, San Diego
- Group of prof. Tajana S. Rosing
- We invented a new machine learning algorithm which can achieve on average 34.1x and 4.1x energy efficiency improvement on training and testing respectively, and 8.2x and 13.4x faster computing in training and testing. This is compared to the previous state-of-the-art HD computing ML algorithms. It provides similar classification accuracy as low-cost state-of-the-art Machine Learning algorithms, while significantly improving the energy efficiency.
2019
SemiHD: Semi-Supervised Learning Using Hyperdimensional Computing
- Conference: ICCAD 2019 - IEEE/ACM International Conference On Computer Aided Design in Colorado
- Work at University of California, San Diego
- Group of prof. Tajana S. Rosing
- Contribution: Hyperdimensional computing applied to semi-supervised machine learning. We noticed an increase in performance, compared to state-of-the-art algorithms.
2019
AdaptHD: Adaptive Efficient Training for Brain-Inspired Hyperdimensional Computing
- Conference: BioCAS 2019 - IEEE Biomedical Circuits and Systems in Japan
- Work at University of California, San Diego and EPFL, Lausanne
- Group of prof. Tajana S. Rosing and Giovanni De Micheli
- AdaptHD introduces the definition of learning rate in HD computing and proposes two approaches for adaptive training: iteration-dependent and data-dependent. In the iteration-dependent approach, AdaptHD uses a large learning rate to speedup the training procedure in the first iterations, and then adaptively reduces the learning rate depending on the slope of the error rate.
2019
CompHD: Efficient Hyperdimensional Computing Using Model Compression
- Conference: ISLPED 2019 - ACM/IEEE International Symposium on Low Power Electronics and Design in Switzerland
- Work at University of California, San Diego
- Group of Prof. Tajana S. Rosing
- Contribution: we developed a model compression techniques for HD computing-based Machine Learning algorithms
2018
Particle shape accounts for instrumental discrepancy in ice core dust size distributions
- Journal: Climate of the Past / European Geosciences Union
- Work at the Niels Bohr Institute, University of Copenhagen
- Group of Prof. Dorthe Dahl-Jensen
- Contribution: significant progress in measurement techniques for ice and climate science
CURRENTLY UNDER REVIEW
Most of the preprints aren't available on arXiv, but you can always contact me for more info!
2019
A Stochastic Acceleration Method for HD Computing-Based Machine Learning
- Submitted to DAC 2019 - IEEE Design Automation Conference
- Work at EPFL
- Group of Prof. Giovanni De Micheli (EPFL)
- Contribution: We developed a new stochastic retraining method for improving the classification accuracy of HD computing-based Machine Learning algorithms
2019
LookHD: Acceleration of Hyperdimensional Computing Exploiting Computation Reuse
- Work at University of California, San Diego
- Group of prof. Tajana S. Rosing
- We invented LookHD: a method which compresses a HD algorithm matrix into a single vector without significantly decreasing classification accuracy. This method is 2.2x faster and 4.1x more energy efficient, as compared to existing HD computing algorithms
DRAFTS
2019
Quantum Computational Finance: Convex Optimization Algorithms for Quantum Computers
- Work at CQT, National University of Singapore and MIT
- Group of Dr. Patrick Rebentrost
- Contribution: We exhibit a quantum algorithm for the pricing of financial assets and derivatives
2019
Bell Diagonal and Werner state generation: entanglement and discord witnesses on the IBM quantum computer
- Work at EPFL
- Contribution: We developed a circuit for creating Bell Diagonal and Werner states
2019
Bit-Serial HD: Approximate Hyperdimensional Computing in Machine Learning
- Work at University of California, San Diego and EPFL
- Groups of Prof. Tajana S. Rosing (UCSD) and Prof. Giovanni De Micheli (EPFL)
- We invented a new HD computing acceleration framework, which significantly reduces the power consumption of Machine Learning algorithms, while maintaining almost the same classification accuracy
2019
Global increase in atmospheric dust content over the last 300 years
- Work at Niels Bohr Institute, University of Copenhagen
- Group of Prof. Dorthe Dahl-Jensen & Prof. Paul Vallelonga
- Contribution: discovery of a large increase in atmospheric dust content, which will have significant impact on our planet's climate
projects
ONGOING PROJECTS
2019
Circuit optimization algorithms for Quantum Computers (Qiskit)
- Implementation of optimization and logic synthesis algorithms for Qiskit-Terra: an open source Python library for Quantum Computers maintained by IBM
- Work in the group of Prof. Giovanni de Micheli and Dr. Mathias Soeken at EPFL
- My pull requests and contributions can be found on GitHub
COMPLETED PROJECTS
2019
Numerical simulations of quantum devices on high performance clusters
- Work at EPFL in the group of prof. Tobias Kippenberg
- Parallelization of simulations on large computing clusters
2018
Text Sentiment Classification
- We developed an algorithm which can analyze the meaning of Tweets
- Using TensorFlow, Keras & GloVe
2018
2016
Simulation of a Tokamak Fusion Reactor
- I created a tokamak fusion reactor simulation in Matlab. It demonstrates how charged high-energy particles can be held in a magnetic bottle (the Tokamak), without touching its walls and melting them due to extremely high temperatures. The magnetic field has a very complicated shape, as it is created by multiple current coils and a rotating charged plasma cloud.
- A simplified version of the code is available on GitHub
2018
