Max Planck - Amazon Science Hub 2024 Research Symposium (Symposium)

This event will be held at the MPI-IS Tübingen Institute on Monday September 16th and at the Amazon Research & Development Building Tübingen on Tuesday September 17th (see below for FAQs).

Science Hub Website

> Detailed Agenda, Abstracts and Bios (pdf) <
 

MONDAY, September 16th

11:45 am   Welcome Remarks
                    Michael J. Black, Managing Director, MPI for Intelligent Systems (MPI-IS)
                    Michael Hirsch, Applied Science Manager, Amazon

12:00 am   Lunch Break

1:00 pm     Keynote Talk
                  Haptic Intelligence
                    Katherine J. Kuchenbecker, Director, Haptic Intelligence department, MPI-IS

1:30 pm    Project Talks 
                 Learning clothing pressure fields via physics-based simulations
                   Gokhan Serhat, Assistant Professor at KU Leuven and Research Scientist at MPI-IS
                   Yashwardhan Deshmukh, Research Software Engineer, Haptic Intelligence department, MPI-IS

                 Physically Plausible Object Pose Refinement in Cluttered Scenes
                   Jörg Stückler, Professor for Intelligent Perception in Technical Systems at the University of Augsburg

                 What do foundation models know about 3D humans? (remote)
                   Gerard Pons-Moll, Professor, Department of Computer Science, University of Tübingen

                 Computer vision for exploring nonverbal human communication
                   Carolin Schmitt, Research Engineer, Optics and Sensing Laboratory, MPI-IS
                   Gökce Ergün, Research Engineer, Optics and Sensing Laboratory, MPI-IS

2:30 pm    Break

2:50 pm    Root Causal Analysis (RCA) at Amazon: some results and challenges
                   Sergio Herna Garrido Mejia, Ph.D. Student, Empirical Inference Department, MPI-IS

                 Improving the Al recycling rate with thermodynamics-guided techniques
                    Waleed Mohammed, Ph.D. Student, Department Microstructure Physics and Alloy Design, MPI for Sustainable Materials

3:20 pm    Talk
                 Physics-Inspired Fit-Aware Virtual Try-On
                   Ming Lin, Amazon Scholar

                 Generative 3D and Novel View Synthesis
                   Sunil Hadap, Principal Applied Scientist Amazon Fashion and Fitness

3:50 pm    Poster Session & Networking

TUESDAY, September 17th

9:30 am     Introduction
                    Betty Mohler Tesch, Principal Applied Scientist Amazon Fashion and Fitness
                    Michael Hirsch, Applied Science Manager Amazon (tbc)

9:45 am     Keynote Talk
                  Adversarial Training Should Be Cast as a Non-Zero-Sum Game
                    Volkan Cevher, Amazon Academic Research Consultant, Amazon Artificial General Intelligence Foundations

10:15 am   Keynote Talk
                  GenAI in the Auto & Manufacturing Industry
                    Chris Markus-Kratz, Director, AWS Automotive

10:35 am   Talks
                  AI in Sports
                     Ianir Ideses, Senior Manager Applied Science Amazon Prime Video

10:55 am    Break

11:15 am    (CANCELLED) Making Time Series Classification Understandable using Multiple Instance Learning
                     Jas Kandola Senior Manager Applied Science Amazon Prime Video

11:35 am    Packaging Optimization at Amazon Scale: Tackling Scientific Challenges with Machine
                     Learning and ML Operational Excellence
                     Matthias Polozek, Principal Applied Scientist Amazon Robotics

11:55 am    REXEL: An end-to-end model for document-level relation extraction and entity linking
                     Shubhi Tyagi, Senior Applied Scientist Amazon Artificial General Intelligence

12:15 pm    Lunch Break

1:15 pm      Generative AI for Ambient Sound: Fusing Text, Melody, and Context for Personalized Experiences
                     Hassen Dhrif, Senior Applied Scientist, Amazon Devices

1:35 pm      Distributed and Adversarial Training of Large Machine Learning Models
                     Michael Mühlebach, Research Group Leader, Learning and Dynamical Systems group, MPI-IS

1:55 pm      Theoretical Foundations of Deep Selective State-Space Models 
                     Antonio Orvieto, Principal Investigator at the ELLIS Institute Tübingen and independent group leader at MPI-IS

2:15 pm     Break

2:35 pm      When do adversarial attacks against language models matter?
                     Jonas Geiping, Principal Investigator at the ELLIS Institute Tübingen and independent research group leader at MPI-IS

2:55 pm      Improving Decision Making with Machine Learning, Provably
                     Manuel Gomez-Rodriguez, Faculty at the MPI for Software Systems

3:15 pm      Generative AI for Research: Global Examples how researchers leverage cloud services to accelerate their work
                      Roberta Piscitelli, Academic Research Lead EMEA
                      Andreas Wagner, Sr. Account Manager for Research in Germany

3:35 pm       Inference optimization for foundational models on AI accelerators
                      Kailash Budhathoki, Sr. Applied Scientist, AWS Deep Engine-Science

FAQs