Program

The research symposium will take place online in its entirety. Attendance is open to all members of Max Planck Institutes. The schedule is shown below.

Please note the following points:

• Times are displayed in Central European Time (CET)
• Lectures will be delivered live on Zoom as per the schedule below.

Wednesday, March 8, 2023

• 3:45 – 4:00 p.m.: Opening remarks, Derek Dreyer, MPI-SWS

Session: Networked Systems

• 4:00 – 4:45 p.m.: Minlan Yu, Harvard University
                                  Rearchitecting the network stack for high performance applications
• 4:45 – 5:30 p.m.: Manya Ghobadi, MIT
                                  Next-Generation Optical Networks for Emerging ML Workloads
• 5:30 – 6:00 p.m.: Break & Discussion

Session: Programming Systems

• 6:00 – 6:45 p.m.: Shan Lu, University of Chicago
                                   Improving the Correctness and Efficiency of Machine Learning    
                                  Applications
• 6:45 – 7:30 p.m.: Miryung Kim, University of California, Los Angeles
                                  Software Engineering for AI
• 7:30 – 8:00 p.m.: Discussion

Thursday, March 9, 2023

Session: Reliable Systems

• 3:15 – 4:00 p.m.: Viktor Vafeiadis, MPI-SWS
                                  Taming weak memory consistency
• 4:00 – 4:45 p.m.: Necmiye Ozay, University of Michigan
                                  Formal methods for Cyber Physical Systems: State of the Art and
                                  Future Challenges
• 4:45 – 5:30 p.m.: Björn Brandenburg, MPI-SWS
                                  PROSA: Trustworthy Schedulability Analysis for Safety-Critical
                                  Real-Time Systems
• 5:30 – 6:00 p.m.: Break & Discussion

Session: Human-Centric Machine Learning

• 6:00 – 6:45 p.m.: Ani Nenkova, Adobe Research
                                  Research Informing Practice in Language Technology
• 6:45 – 7:30 p.m.: Manuel Gomez Rodriguez, MPI-SWS
                                  Improving Decision Making with Machine Learning, Provably
• 7:30 – 8:00 p.m.: Discussion

Friday, March 10, 2023

Session: Security

• 4:00 – 4:45 p.m.: Catalin Hritcu, MPI-SP
                                   Formally Secure Compilation of Unsafe C Compartments
• 4:45 – 5:30 p.m.: Alina Oprea, Northeastern University
                                   On the connection between integrity and privacy attacks in
                                   adversarial machine learning
• 5:30 – 6:00 p.m.: Break & Discussion

Session: Human-Centric Computing & Robotics

• 6:00 – 6:45 p.m.: Katharina Reinecke, University of Washington
                                   Biases in Everyday Digital Technology and How to Address Them
• 6:45 – 7:30 p.m.: Anca Dragan, University of California, Berkeley
                                   Robotics algorithms that take people into account
• 7:30 – 8:00 p.m.: Discussion