Leon Sering

Leon Sering

Combinatorial Optimization Expert & Postdoctoral Researcher

ETH Zürich - Institute for Operations Research

Hey, I’m Leon!

As a mathematician and computer scientist, I currently work as a Postdoc at the Institute for Operations Research at ETH Zürich. Beside my theoretical research on Nash flows over time (agent-based traffic modeling), my true passion lies in optimization algorithms and high performance algorithm design.

At the moment I am in the lead of two exciting industry collaborations:

  • With the Swiss Post, we optimize the tours for same-day delivery in Zurich and other Swiss cities using local-search-based meta-heuristics.

  • For the Swiss Federal Railways SBB, I’m developing a high performance algorithm for optimal rolling stock scheduling.

My programming language of choice? Rust. This high performance system programming language, with focus on fearless concurrency, is the ideal tool for my projects. And most importantly: It’s fun to use!

In my downtime, you will find me (together with my family) exploring the wonders around the world – through road trips, photography, and scuba diving.

Welcome to my website!

Here, I showcase my projects, my scientific work, and our travel adventures.

Interests
  • Combinatorial Optimization
  • Operations Research
  • High Performance Computing
  • Meta-Heuristics
  • Algorithm Design
  • Parallelism/Concurrency
  • Rust (Programming Language)
Education

  • Dr. rer. nat. in Mathematics (PhD), 2020

    Technical University of Berlin

  • Master of Advanced Studies in Mathematics, 2014

    University of Cambridge

  • Master of Science - Mathematics, 2016

    University of Würzburg

  • Bachelor of Science - Mathematics, 2013

    University of Würzburg

  • Bachelor of Science - Computer Science, 2013

    University of Würzburg

Experience

 
 
 
 
 
Optano GmbH
Operations Research Expert (Freelancer)
Optano GmbH
May 2023 – June 2023 Paderborn, Germany

Consulting services for a line haul solution:

  • design of mixed-integer linear program
  • implementation of a fast multi-objective routing algorithm
 
 
 
 
 
ETH Zürich
Postdoctoral Researcher
ETH Zürich
April 2021 – Present Zurich, Switzerland

Institute for Operations Research, Department of Mathematics:

  • leading scientific collaboration with Swiss Post and Swiss Federal Railways SBB
  • design and implementation of a fast clustering algorithm
  • research in efficient algorithms, meta-heuristics, parallelism, and operations research
 
 
 
 
 
Technical University of Berlin
Research Associate in Discrete Mathematics
Technical University of Berlin
September 2016 – March 2021 Berlin, Germany

Research Group: Combinatorial Optimization & Graph Algorithm, Institute of Mathematics:

  • ECMath and MATH+ research projects: dynamic models and algorithms for equilibria in traffic networks
  • research goals: improving agent-based mathematical flow over time models (Nash flow over time) to connect with large-scale traffic simulations such as MATSim

Skills

Technical
Combinatorial Optimization
Meta-Heuristics
Fast Parallel Algorithms
Mathematical Programming
Traffic / Logistic Optimization
Programming
Rust
Scala
C++
Python
Java
Tools
Git / GitLab / GitHub
NeoVim
Linux / Bash
Gurobi / SCIP
Docker
Hobbies
Traveling / Road Trips
Scuba Diving
Computer Games / Virtual Reality
Photography
Running

Projects

*
All Major Minor Optimization
Rolling Stock Scheduling Optimization
A collaboration between Swiss Federal Railways SBB and ETH Zürich.
Code
Rolling Stock Scheduling Optimization
Fast Same-Day Delivery Optimization
A collaboration between Swiss Post / Notime and ETH Zürich.
Fast Same-Day Delivery Optimization
Fair and Fast k-Center Clustering
A linear-time k-center algorithm with fairness conditions and worst-case guarantees that is very fast in practice. Written in Rust with Python bindings.
Code Doc Crate Publication
Fair and Fast k-Center Clustering
Nash Flow Computation
A GUI-tool for computing Nash flows over time with or without spillback as well as (spillback) thin flows.
Code
Nash Flow Computation
Axonium
Automatic measurements of axon lengths, the nerve fibers connecting neurons in a brain.
Code
Axonium
Magic Eye
A small Python-script to create an autostereogram.
Code
Magic Eye
Focus Region Tool
A tool to enlarge a focus region in a road map while minimizing the distortion by convex quadratic programming.
Code Publication
Focus Region Tool
BWInf - Career Information Day Scheduling
Submission for the second round of the German National Competition in Computer Science in 2007/2008.
Code
BWInf - Career Information Day Scheduling

Dissertation

Nash Flows Over Time
Motivated by the dynamic traffic assignment problem and with the goal in mind to obtain a better understanding of the complex traffic …
Leon Sering
PDF Cite DOI
Nash Flows Over Time

Recent Publications

Convergence of Approximate and Packet Routing Equilibria to Nash Flows Over Time
We consider a dynamic model of traffic that has received a lot of attention in the past few years. Infinitesimally small agents aim to …
Neil Olver, Leon Sering, Laura Vargas Koch
PDF Cite DOI arXiv
Convergence of Approximate and Packet Routing Equilibria to Nash Flows Over Time
ReLU Neural Networks of Polynomial Size for Exact Maximum Flow Computation
This paper studies the expressive power of artificial neural networks with rectified linear units. In order to study them as a model of …
Christoph Hertrich, Leon Sering
PDF Cite DOI arXiv
ReLU Neural Networks of Polynomial Size for Exact Maximum Flow Computation
Spillback changes the long-term behavior of dynamic equilibria in fluid queuing networks
We study the long-term behavior of dynamic traffic equilibria and find that it heavily depends on whether spillback is captured in the …
Theresa Ziemke, Leon Sering, Kai Nagel
PDF Cite DOI
Spillback changes the long-term behavior of dynamic equilibria in fluid queuing networks
Atomic Splittable Flow Over Time Games
In an atomic splittable flow over time game, finitely many players route flow dynamically through a network, in which edges are …
Antonia Adamik, Leon Sering
PDF Cite DOI arXiv
Atomic Splittable Flow Over Time Games
Continuity, Uniqueness and Long-Term Behavior of Nash Flows Over Time

We consider a dynamic model of traffic that has received a lot of attention in the past few years. Users control infinitesimal flow …

Neil Olver, Leon Sering, Laura Vargas Koch
PDF Cite DOI arXiv
Continuity, Uniqueness and Long-Term Behavior of Nash Flows Over Time
See all publications

Traveling

USA 2022
Over a period of six months, my girlfriend and I road-tripped across the stunning landscapes of the United States, exploring various national and state parks in our Jeep Liberty named “Rusty”.
USA 2022
Dolomites 2019
In a 4-day hiking adventure through the Sexten Dolomites, my girlfriend and I traversed the Elferscharte to the Zsigmondyhütte, circled the iconic Drei Zinnen amid stunning alpine landscapes.
Dolomites 2019
Namibia 2018
In Namibia and Botswana, my girlfriend and I explored Fish River Canyon, Sossusvlei, and Etosha National Park, marveled at Victoria Falls, and enjoyed a mokoro trip in the Okavango Delta.
Namibia 2018
Australia 2016
In 2016, my girlfriend and I completed Australia’s “Big Lap,” circling the coast from Sydney, exploring iconic landmarks like the Great Ocean Road, Karijini NP, Gibb River Road, and Uluru.
Australia 2016

Contact

Feel free to reach out for collaborative opportunities, academic discussions, or anything related to operations research, combinatorial optimization, parallel algorithm design, or Rust. I look forward to engaging with fellow enthusiasts and professionals in the field!