create your own web page
Mobirise


Lucas Liebenwein

Passionate about machine learning, robotics, and the impact of science on improving people's lives.

I am a PhD candidate at the Computer Science and Artificial Intelligence Lab at MIT under the supervision of Prof. Daniela Rus. My research interests lie at the intersection of deep learning and robotics with applications to autonomous systems such as self-driving cars.

In the future, I envision us to rely on machines and tools powered by deep learning to help us solve complex tasks in our homes, factories, and workplaces. Those algorithms will need to be efficient, interpretable, and above all safe in the sense that they work reliably and consistently in uncertain, complex environments. To this end, my work focuses on developing more efficient, more reliable, and more interpretable deep learning algorithms by combining techniques ranging from probabilistic algorithms, such as coresets and statistical learning theory, to practical tools, such as neural network pruning and data interpretation methods.

Publications

A complete list of my publications can also be found on my Google Scholar!

  1. Provable Filter Pruning for Efficient Neural Networks
    Lucas Liebenwein, Cenk Baykal, Harry Lang, Dan Feldman, Daniela Rus. In International Conference on Learning Representations, 2020.
    [PDF], [BibTex], [Link]
  2. Machine Learning-based Estimation of Forest Carbon Stocks to increase Transparency of Forest Preservation Efforts
    Björn Lütjens, Lucas Liebenwein, Katharina Kramer. In NeurIPS Workshop Tackling Climate Change with Machine Learning, 2019.
    [PDF], [BibTex], [Link]
  3. SiPPing Neural Networks: Sensitivity-informed Provable Pruning of Neural Networks
    Cenk Baykal*, Lucas Liebenwein*, Igor Gilitschenski, Dan Feldman, Daniela Rus. In arXiv preprint arXiv:1910.05422, 2019.
    [PDF], [BibTex], [Link]
  4. Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds
    Cenk Baykal*, Lucas Liebenwein*, Igor Gilitschenski, Dan Feldman, Daniela Rus. In International Conference on Learning Representation, 2019.
    [PDF], [BibTex], [Link]
  5. Counterexample-Guided Safety Contracts for Autonomous Driving
    Jonathan A DeCastro*, Lucas Liebenwein*, Cristian-Ioan Vasile, Russ Tedrake, Sertac Karaman, and Daniela Rus. Counterexample-Guided Safety Contracts for Autonomous Driving. In International Workshop on the Algorithmic Foundations of Robotics (WAFR), 2018.
    [PDF], [BibTex], [Link]
  6. Sampling-Based Approximation Algorithms for Reachability Analysis with Provable Guarantees
    Lucas Liebenwein*, Cenk Baykal*, Igor Gilitschenski, Sertac Karaman, Daniela Rus. Sampling-Based Approximation Algorithms for Reachability Analysis with Provable Guarantees. In Robotics: Science and Systems XIV (RSS), 2018. 
    [PDF], [BibTex], [Link]
  7. Compositional and Contract-Based Verification for Autonomous Driving on Road Networks
    Lucas Liebenwein, Wilko Schwarting, Cristian-Ioan Vasile, Jonathan DeCastro, Javier Alonso-Mora, Sertac Karaman, Daniela Rus. Compositional and Contract-Based Verification for Autonomous Driving on Road Networks. In International Symposium on Robotics Research (ISRR), 2017.
    [PDF], [Video], [BibTex], [Link]

    *equal contribution

Theses

  1. Contract-Based Safety Verification for Autonomous Driving
    Lucas Liebenwein. Contract-Based Safety Verification for Autonomous Driving. S.M. Thesis, Massachusetts Institute of Technology, 2018.
    [PDF], [BibTex], [Link]
  2. Autonomous Pairing of Distributed Flight Array Modules
    Lucas Liebenwein. B.Sc. Thesis, Swiss Federal Institute Of Technology, Zurich, 2015.
    [PDF], [BibTex]

Teaching

  • Graduate Teaching Assistant - "6.856 Randomized Algorithms", MIT, Jan 2019 - June 2019.
    This course is a graduate-level introduction to algorithms that use randomization for combinatorial and sketching algorithms taught by Prof. David Karger. My responsibilities included holding weekly recitations, developing homework problems, and administering the grading process.
  • Co-Author for Course Book - "Mechanics 3 - Dynamics", ETH Zurich, Jun 2015 - Nov 2015.
    Under the guidance of Prof. George Haller, I wrote the course text that accompanies the lecture "Mechanics 3 - Dynamics". This class is taken by approximately 400 Mechanical Engineering students each year during the 2nd year of their Bachelor's.
  • Teaching Assistant - "Mechanics 3 - Dynamics", ETH Zurich, Aug 2014 - Dec 2014.
    I was responsible for a weekly recitation session of 100 students. Together with two other teaching assistants, we held weekly theory recap and Q&A sessions. We also took the initiative to create written summaries of the material for each week.
  • Teaching Assistant - "Computer Science for Engineers I", ETH Zurich, Feb 2014 - Jul 2014.
    I taught a weekly recitation and exercise session. This course is taken by approximately 600 Mechanical Engineering students during their 1st year. Students learn the basics of programming in C/C++ and are exposed to an introduction in algorithms.

Experience

  • Graduate Research Assistant - MIT CSAIL, Prof. Daniela Rus, Sep 2016 - Present.
    My research at the Distributed Robotics Lab, MIT CSAIL evolves around advancing the state-of-the-art in deep learning and as a consequence the autonomy of robots. In particular, I am interested in how we can develop more efficient, more interpretable, and safer deep learning algorithms. This will expand the horizon of potential applications of deep learning in safety-critical domains, such as robotics. Previously, I worked on safety guarantees for autonomous cars in dynamic, real-world settings.
  • Autopilot Software Intern - Tesla Inc., Palo Alto, CA, June 2019 - Sept 2019.
    During my internship at Tesla, I closely worked with the motion planning team on developing the next generation planning algorithms for Tesla Autopilot. I spearheaded the development of a new active safety feature that enhanced the general awareness of the planner with regards to other cars on the road.
  • Visiting Researcher - Singapore-MIT Alliance for Research & Technology Centre, Jan 2017.
    I visited the National University of Singapore (NUS) for a research collaboration and technology transfer between the Future Urban Mobility Group and our research group. 
  • Autonomous Car Engineer - nuTonomy Singapore, Dec 2015 - Feb 2016.
    I worked for nuTonomy, an autonomous car start-up located in Singapore, Boston, and Zurich, tackling the most difficult challenge in autonomous driving: urban driving. During my time, we developed the initial prototype, which would become the foundation for the later expansion of the company and the car fleet. My main responsibilty was to develop, implement, and supervise the automated cruise control for an autonomous vehicle.
  • Undergraduate Research Assistant - ETH, Prof. Raffaello D'Andrea, Sep 2014 - Oct 2015.
    I was an undergraduate research assistant at the Institute for Dynamics Systems and Control (IDSC) and my advisor was Prof. Raffaello D'Andrea. My research was centered around the Distributed Flight Array (DFA), a novel flying platform that consists of multiple units, each capable of driving and flying. I co-lead the implementation of a real-time operating system for the DFA, and developed novel algorithms that allowed the DFA units to self-assemble on the ground.

Projects

Mobirise

Autonomous Driving

The Toyota-CSAIL Joint Research Center is aimed at furthering the development of autonomous vehicle technologies.
Learn more...

Mobirise

Coresets

The goal of this projects is to design novel data compression techniques to accelerate popular machine learning algorithms.
Learn more...

Mobirise

Gentou

A smart bottle cage that acts as a personal coach to coach professional cyclists how to stay hydrated during races.
Learn more...

Mobirise

Distributed Flight Array

A distributed flying platform that is able to drive, dock with their peers, and fly in a coordinated fashion.
Learn more...

Curriculum Vitae

© Copyright 2020 Lucas Liebenwein- All Rights Reserved