Prof. Niki Kilbertus is an Assistant Professor at the Technical University of Munich (TUM) in the Department of Informatics, and Group Leader at Helmholtz AI. His research focuses on causal machine learning, ethical AI systems, and applications in healthcare, climate science, and dynamical systems. He earned his PhD from the University of Cambridge (2020) and has held positions at DeepMind, Google, and Amazon during his studies. His research interests include causal discovery, fairness in AI, counterfactual reasoning, and integrating physics-based constraints into neural networks. Key contributions include foundational work on fair machine learning (e.g., avoiding discrimination through causal models) and developing methods for causal inference in complex systems like healthcare and climate modeling. Recent work emphasizes generative models for causal interventions, robust treatment effect estimation, and physically consistent neural differential equations. He leads a large interdisciplinary group with over 20 students and postdocs working on projects funded by Helmholtz Association, ERC, and industry collaborations. Notable awards include the Leopoldina Prize for Young Scientists (2024) and membership in the Junge Akademie. His lab maintains active partnerships with ELLIS, MCML, and the Zuse Institute Berlin.
Michael Carbin is the Jamieson Career Development Assistant Professor of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology (MIT) and leads the MIT Programming Systems Group. His research focuses on programming systems that address system uncertainty to enhance performance, energy efficiency, and resilience, particularly in environments involving neural networks , approximate computing , and unreliable hardware . His work spans probabilistic programming , quantum computing , and machine learning systems . Articles highlight contributions in pruning neural networks , quantum data structures , and compiler optimization , reflecting trends in deep learning , formal verification , and language-driven systems . Scientific Awards : MIT Frank E. Perkins Award (2020) Sloan Research Fellowship (2020) Facebook Research Award (2019) NSF CAREER Award (2018) Best Paper Awards at OOPSLA (2013, 2014) He has advised numerous graduate students and postdocs including Eric Atkinson, Cambridge Yang, and Charles Yuan, and served on program committees for conferences like POPL, OOPSLA, and ICLR. His group collaborates with institutions such as MIT CSAIL and explores applications in quantum algorithms and probabilistic inference .
Celestine Mendler-Dünner is a Principal Investigator at the ELLIS Institute in Tübingen, co-affiliated with the Max Planck Institute for Intelligent Systems and the Tübingen AI Center. She leads the Algorithms and Society research group, focusing on machine learning in social contexts and the role of prediction in digital economies. Her work bridges theoretical machine learning with practical societal impact, developing tools for safe, reliable, and equitable AI ecosystems. Her educational background includes a PhD from ETH Zurich in collaboration with IBM Research, followed by an SNSF postdoctoral fellowship at UC Berkeley hosted by Moritz Hardt. She was previously a group leader at the Max Planck Institute for Intelligent Systems before joining the ELLIS Institute. Mendler-Dünner's research spans several interconnected themes including performative prediction (where predictions change the behavior they aim to predict), algorithmic collective action (how participants can steer AI systems toward common goals), and the role of LLMs in social science research. Her work combines theoretical foundations with practical implementations, addressing challenges in interactive machine learning, optimization in dynamic environments, and context-specific evaluation of AI systems. She particularly examines how algorithmic predictions mediate services and platforms at societal scale, exploring concepts of economic power in digital markets. Her publication record shows a clear evolution from system-aware machine learning algorithms (including foundational work on IBM Snap ML) toward increasingly sociotechnical questions at the intersection of machine learning, economics, and policy. Recent work focuses on measuring performative power in digital economies, evaluating LLMs as risk scores, and developing frameworks for algorithmic collective action in recommender systems and labor markets. Among her notable recognitions are the ETH Medal for her dissertation, the IBM Research Division Award, the Fritz Kutter Award, and the IBM Eminence and Excellence Award. She is an ELLIS Scholar, a fellow of the Elisabeth-Schiemann-Kolleg, and affiliated with several prestigious research programs including the International Max Planck Research School for Intelligent Systems and the Max Planck ETH Center for Learning Systems. ETH Medal (dissertation award) IBM Research Division Award Fritz Kutter Award IBM Eminence and Excellence Award SNSF Early Postdoc Mobility Fellowship Mendler-Dünner actively mentors the next generation of researchers, advising PhD student Patrik Wolf and supervising research interns including Joachim Baumann, Haiqing Zhu, and Anna Badalyan, as well as Master's student Dorothee Sigg. She serves as core faculty for the International Max Planck Research School and associated faculty for the Max Planck ETH Center for Learning Systems. Her group has secured significant research funding through fellowships and institutional support, enabling work on projects like Powermeter (measuring search engine influence) and Snap ML (resource-efficient machine learning library with over 1 million PyPI downloads). She leads the Algorithms and Society research group, which examines machine learning as part of broader sociotechnical ecosystems. The group explores human-population interactions with algorithmic systems and incorporates these insights into learning system fundamentals. Current projects include investigating economic incentives in digital platforms, developing tools for systematic LLM evaluation in social science contexts, and creating frameworks for collective action in algorithmic systems. Mendler-Dünner also co-organizes the Algorithmic Collective Action workshop at NeurIPS 2025, demonstrating her leadership in emerging research directions at the AI-society interface.
Florian Tramèr is an Assistant Professor in the Department of Computer Science at ETH Zurich, Switzerland, leading research at the intersection of machine learning security, privacy, and AI safety. His work focuses on identifying and mitigating security vulnerabilities in machine learning systems, particularly in large language models and other AI systems. Tramèr's primary research interests include adversarial machine learning, membership inference attacks, privacy-preserving AI, and the security implications of large language models. His work has significantly advanced our understanding of how machine learning models memorize training data, how this memorization creates privacy risks, and how to evaluate the robustness of machine learning systems against various attacks. His recent publications demonstrate a strong focus on practical security challenges in deployed AI systems, including data extraction from language models, adversarial attacks against generative AI, and developing more rigorous evaluation methodologies for machine learning security. Tramèr's research has been published in top venues including ICLR, NeurIPS, ICML, and IEEE Security & Privacy. Tramèr is actively collaborating with leading researchers in the field including Nicholas Carlini, Matthew Jagielski, and Javier Rando, contributing to important initiatives like the International AI Safety Report. His work bridges theoretical security concepts with practical implications for real-world AI deployment.
Prof. Matthias Nießner is a Professor at the Technical University of Munich , where he leads the Visual Computing Lab . Prior to this, he held a Visiting Assistant Professor position at Stanford University . His work bridges computer vision , graphics , and machine learning , focusing on 3D reconstruction , semantic scene understanding , and AI-driven video synthesis . Prof. Nießner has published over 150 works in top venues like SIGGRAPH , CVPR , and ECCV , with several receiving best paper awards (SIGCHI’14, HPG’15, SPG’18, SIGGRAPH’16 Emerging Tech). His research has garnered international media attention, including features in the New York Times , Wall Street Journal , and MIT Technological Review , as well as TV demonstrations (e.g., Jimmy Kimmel Live for Face2Face technology). Awards : TUM-IAS Rudolph Moessbauer Fellowship (2017–ongoing) Google Faculty Award (2017) Nvidia Professor Partnership Award (2018) ERC Starting Grant (2018, €1.5M) Eurographics Young Researcher Award (2019) Research Trends : 3D Gaussian Splatting for real-time rendering Neural Radiance Fields (NeRF) with mesh supervision Audio-driven facial animation via diffusion models Latent space diffusion for 3D scenes Self-supervised and zero-shot methods for 3D and image analysis As a co-founder and director of Synthesia Inc. , he drives democratization of synthetic media. His YouTube channel has over 5 million views, reflecting his impact beyond academia.
Jonathan T. Barron is a Researcher at Google DeepMind in San Francisco, specializing in Computer Vision , Neural Rendering , and 3D Scene Reconstruction . He earned his PhD at UC Berkeley under Jitendra Malik and has pioneered advancements in NeRF (Neural Radiance Fields) and diffusion-based 3D generation. Research Interests : Computer Vision, Deep Learning, Generative AI, Image Processing, and 3D Reconstruction via Radiance Fields. His work includes Bolt3D for rapid 3D scene generation, CAT3D/CAT4D for text-to-3D/4D, and Zip-NeRF for anti-aliased radiance fields. He has also developed real-time rendering frameworks like SMERF and NeRF-Casting for reflections. Scientific awards: PAMI Young Researcher Award He has served as Area Chair for CVPR, ICCV, and NeurIPS, and his research is widely adopted in applications like Google's Lens Blur , Portrait Mode , and Jump VR .
Yin Tat Lee is an Associate Professor at the Paul G. Allen School of Computer Science & Engineering , University of Washington, and a Senior Principal Researcher in Microsoft AI. His research spans convex optimization , convex geometry , graph algorithms , online algorithms , and differential privacy , with applications in machine learning and theoretical computer science.
Zeynep Akata is the Liesel Beckmann Distinguished Professor of Computer Science at the Technical University of Munich (TUM) and Director of the Institute for Explainable Machine Learning at Helmholtz Munich. Previously she was a W3 Professor at the University of Tübingen (2019-2023) and held faculty and post-doctoral positions at the University of Amsterdam, UC Berkeley and the Max Planck Institute for Informatics. Her research focuses on multimodal learning and explainable artificial intelligence . Education: PhD, University of Grenoble / INRIA Rhône-Alpes, 2014 MSc, RWTH Aachen University, 2010 BSc, Trakya University, Turkey, 2008 Research Interests: Professor Akata’s group develops algorithms that learn from vision, language and other modalities simultaneously, with a strong emphasis on zero-shot, few-shot and continual learning . A central theme is making decisions interpretable, leading to work on explainable AI, concept bottleneck models, multimodal reasoning and human-aligned representation learning . Recent projects investigate large-scale multimodal language models, dataset distillation, model merging and continual knowledge editing. Publication Trends: Her 2024-2025 publications reveal a shift toward foundational large-scale models (diffusion, LLMs, vision-language transformers) while retaining the core themes of interpretability and generalization under limited supervision . Topics span dataset distillation, model merging, continual learning, fairness auditing of generative models and novel evaluation protocols for zero-shot learning systems. Scientific Awards: Lise-Meitner Award for Excellent Women in Computer Science (2014) Young Scientist Honour, Werner-von-Siemens-Ring Foundation (2019) ERC Starting Grant, European Commission (2019) DAGM German Pattern Recognition Award (2021) ECVA Young Researcher Award (2022) Alfried Krupp Award (2023) Advising & Funding: Prof. Akata currently supervises or co-supervises 25+ PhD students across TUM and the University of Tübingen via ELLIS and IMPRS-IS doctoral programs. She holds major grants including an ERC Starting Grant and DARPA Explainable AI funding, and is a frequent program chair and area chair for premier conferences (CVPR 2024, ECCV 2026, NeurIPS, ICML, etc.). Labs & Teams: She leads the Institute for Explainable Machine Learning at Helmholtz Munich and heads the Multimodal Learning and Explainable AI group at TUM. The institute collaborates closely with the ELLIS Institute Tübingen and Cyber Valley ecosystem, and maintains close ties with the Max Planck Institute for Intelligent Systems and Informatics.
Chita R. Das is a Professor at Pennsylvania State University, known for extensive contributions in computer architecture, machine learning, and high-performance computing. Their research focuses on optimizing hardware-software co-design for edge computing, cloud infrastructure, and energy-efficient systems. Key areas include FPGA acceleration, GPU optimization, and serverless computing frameworks. Das collaborates frequently with institutions like AMD and Intel, addressing challenges in parallel computing and distributed systems. Their work bridges theoretical advancements with practical applications in recommendation systems, bioinformatics, and real-time video processing. Research interests span across hardware acceleration techniques, cloud resource management, and sustainable computing. Notable projects include adaptive training frameworks for intermittent power environments and neural-augmented game streaming for mobile platforms. Das's publications often address performance bottlenecks in modern architectures and propose novel solutions for latency and energy efficiency. Recent articles highlight innovations in serverless computing cost optimization, low-bandwidth VR streaming, and FPGA-based bioinformatics tools. Their contributions are characterized by interdisciplinary approaches combining computer architecture with machine learning and embedded systems.
Matthias Feurer is a Thomas Bayes Fellow and interim professor at the Chair of Statistical Learning and Data Science, funded by the Munich Center for Machine Learning (MCML) at Ludwig Maximilian University of Munich. He is a member of the Department of Statistics at LMU Munich, working under Prof. Dr. Bernd Bischl. His academic background includes: PhD in Computer Science from Albert-Ludwigs-Universität Freiburg, supervised by Prof. Dr. Frank Hutter M.Sc. in Computer Science from the University of Freiburg B.Sc. in Computer Science and Media from the Media University Stuttgart Feurer's research focuses on simplifying machine learning usage through Automated Machine Learning (AutoML). His work encompasses hyperparameter optimization, meta-learning, and model selection, with increasing emphasis on multi-objective AutoML that considers factors beyond predictive performance such as interpretability, deployability, and fairness. He actively develops open-source tools to advance the field. His recent publications demonstrate a strong trajectory in practical AutoML systems, with growing attention to tabular machine learning, foundation models integration, and addressing real-world constraints in optimization. His work consistently bridges theoretical advances with practical implementations through several widely-used open-source projects. Notable achievements include: 1st place in the warmstarting-friendly leaderboard of the BBO NeurIPS challenge Winner of the 2nd AutoML challenge Winner of the kdnuggets blog contest on AutoML Feurer is actively mentoring and teaching, having advertised PhD positions focused on AutoML, optimization, and benchmarking. He co-founded the Open Machine Learning Foundation supporting OpenML.org. His upcoming move to TU Dortmund as an assistant professor in AutoML and Optimization signals continued growth in his academic career while maintaining his research focus on making machine learning more accessible and rigorous.
Antonia Holzapfel is a doctoral researcher at the Institute for Data Science in Mechanical Engineering (DSME) of RWTH Aachen University. Supervised by Prof. Sebastian Trimpe, her work focuses on safe and explainable machine learning (XAI) for robotics and time-series modeling, with applications in mechanical engineering and medical technology. Bachelor's in Mechanical Engineering (Design Engineering specialization) Master's in General Mechanical Engineering (Simulation Technology and Medical Technology specializations) Her research addresses safety in learning processes , including Bayesian optimization for quadcopters and concept extraction for model interpretability. Her Friedrich-Wilhelm Award -winning thesis (2024) explored safe online learning in time-varying environments, presented at the L4DC conference. She contributes to DFG-funded projects on data-driven process modeling in forming technology. Key scientific contributions include: Advancing safe Bayesian optimization for autonomous systems Developing explainable AI methods for time-series analysis Improving model robustness through internal representation patterns Scientific awards: Friedrich-Wilhelm Award for outstanding Master's thesis (2024) Antonia's work bridges machine learning safety with practical applications in robotics and industrial processes.
Dr. Vincent Fortuin is a tenure-track Assistant Professor at the Technical University of Munich (TUM) and a research group leader at Helmholtz AI in Munich. He leads the Efficient Learning and Probabilistic Inference for Science (ELPIS) group and holds multiple prestigious fellowships including the Branco Weiss Fellowship. His academic affiliations include the TUM School of Computation, Information and Technology, the Konrad Zuse School of Excellence in Reliable AI, and the Munich Center for Machine Learning. Dr. Fortuin earned his BSc in Molecular Life Sciences from the University of Hamburg (2012-2015), followed by an MSc in Computational Biology and Bioinformatics from ETH Zürich (2015-2017), where he received the ETH Excellence Scholarship and the Willi Studer Prize. He completed his PhD in Machine Learning at ETH Zürich (2017-2021) under the supervision of Gunnar Rätsch and Andreas Krause, supported by a Swiss Data Science Center PhD Fellowship. Prior to joining TUM, he was a Research Fellow at St. John's College, University of Cambridge (2022-2023). His research focuses on the intersection of Bayesian statistics and deep learning, specifically developing methods for more robust, data-efficient AI systems with reliable uncertainty estimates. His work addresses critical limitations in standard deep learning approaches, particularly their tendency to be overconfident in predictions and require large datasets for training. He investigates better priors and more efficient inference techniques for Bayesian deep learning, deep generative modeling, meta-learning, and PAC-Bayesian theory, with applications in scientific and biomedical domains. Dr. Fortuin's recent publications demonstrate a consistent focus on improving uncertainty quantification in deep learning systems, with increasing emphasis on practical applications in scientific contexts. His work spans from theoretical foundations of Bayesian deep learning to practical implementations in protein design, materials science, and medical applications. A notable trend is his exploration of how to make Bayesian methods more scalable and applicable to modern large-scale AI systems while maintaining theoretical guarantees. Branco Weiss Fellowship (2023) St John's College Research Fellowship (2022) Swiss National Science Foundation Postdoc.Mobility Fellowship (2022) Swiss Data Science Center PhD Fellowship (2018) ETH Excellence Scholarship (2015) Willi Studer Award (2018) Dr. Fortuin actively supervises PhD and Master's students through his ELPIS research group at Helmholtz AI. He serves as a regular reviewer and area chair for major machine learning conferences and is an action editor for TMLR. He co-organizes the Symposium on Advances in Approximate Bayesian Inference (AABI) and the ICBINB initiative, demonstrating his commitment to advancing the field through community building. His research group receives funding from multiple sources including Helmholtz AI, the Branco Weiss Fellowship, and collaborations with international institutions. Dr. Fortuin leads the Efficient Learning and Probabilistic Inference for Science (ELPIS) group at Helmholtz AI, which focuses on fundamental machine learning research motivated by real-world scientific problems. The group collaborates extensively with researchers across Helmholtz centers and international institutions, particularly in biomedical applications where reliable uncertainty estimates are crucial.
Carlos Cinelli is an Assistant Professor in the Department of Statistics at the University of Washington, where he conducts research at the intersection of causal inference, statistical methodology, machine learning, and artificial intelligence. He is also a data science fellow at the eScience Institute and affiliate faculty of the Center for Statistics and the Social Sciences, demonstrating his interdisciplinary approach to causal methodology. Dr. Cinelli received his Ph.D. in Statistics from the University of California, Los Angeles, advised by Chad Hazlett and Judea Pearl, two prominent figures in causal inference. His research focuses on developing new causal and statistical methods for transparent and robust causal claims in empirical sciences, with particular attention to challenges faced by social and health scientists. His work spans theoretical developments in causal identification, sensitivity analysis frameworks, and practical software implementations that enable researchers to assess the robustness of their causal conclusions. Cinelli's research program addresses fundamental questions about how unobserved confounding affects causal estimates and develops tools to quantify how sensitive findings are to potential violations of causal assumptions. His work on omitted variable bias frameworks has been particularly influential across multiple disciplines. Through his publications, Cinelli has established himself as a leading researcher in causal inference methodology, with papers appearing in top journals across statistics, machine learning, epidemiology, and social sciences. His work demonstrates both theoretical rigor and practical relevance, often accompanied by open-source software implementations that make his methods accessible to applied researchers. Best paper award at SBE 2024 in Econometrics Royalty Research Fund (RRF) Award recipient NSF/MMS research support As an advisor, Cinelli has successfully guided PhD students like Nick Irons to dissertation completion. He actively seeks new students with strong interests in causal inference. His research is supported by multiple funding sources including the National Science Foundation and the University of Washington's Royalty Research Fund. Cinelli contributes to the academic community through editorial work for the Journal of Causal Inference and by developing widely used software packages like sensemakr for sensitivity analysis.
Michael Muehlebach leads the independent Learning and Dynamical Systems research group at the Max Planck Institute for Intelligent Systems in Tuebingen, Germany. His interdisciplinary work bridges machine learning, dynamical systems theory, and control engineering to develop algorithms for cyber-physical systems with theoretical guarantees and practical implementations. Dr. Muehlebach received his B.Sc. and M.Sc. in Mechanical Engineering from ETH Zurich in 2010 and 2013, specializing in robotics and control systems. He completed his Ph.D. at ETH's Institute for Dynamic Systems and Control under Prof. R. D'Andrea in 2018, followed by postdoctoral research with Prof. Michael I. Jordan at UC Berkeley. His research focuses on constrained optimization, reinforcement learning, and control theory with applications in robotics. He pioneered approaches that express constraints in terms of velocities rather than positions, enabling more efficient optimization algorithms. His work spans theoretical foundations to physical implementations, including the One-Wheel Cubli balancing robot and electromagnetic navigation systems. Recent publications reveal a strong trend toward physics-informed machine learning, particularly for robotics applications requiring real-time performance and safety guarantees. Dr. Muehlebach has received numerous prestigious awards: Outstanding D-MAVT Bachelor Award Willi-Studer prize for best Master's degree ETH Medal and HILTI prize for doctoral thesis Branco Weiss Fellowship (2018) Emmy Noether Fellowship (2020) Amazon Fellowship (2024) He actively mentors doctoral researchers including Hao Ma, Melis Ilayda Bal, and Onno Eberhard, with research supported by multiple grants. His group maintains strong collaborations with Bernhard Schölkopf's Empirical Inference group at the Max Planck Institute. The Learning and Dynamical Systems group develops innovative hardware and software platforms, including Floaty (a wind-harnessing flying robot), advanced electromagnetic navigation systems, and data-efficient learning methods for robotic table tennis. Their approach combines rigorous theoretical analysis with practical validation on physical systems, emphasizing the integration of known physical structure into machine learning algorithms to improve sample efficiency and ensure generalization.
Tianmin Shu is an Assistant Professor in the Department of Computer Science at Johns Hopkins University , with a joint appointment in the Department of Cognitive Science . He is the founding director of the Social Cognitive AI (SCAI) Lab and was previously a Research Scientist at MIT, working with Josh Tenenbaum and Antonio Torralba. His research goal is to advance human-centered AI by engineering machine social intelligence —building systems that understand, reason about, and interact with humans in real-world settings. His work is inherently interdisciplinary, integrating machine learning, computer vision, robotics, and social cognition. Research Interests: Theory of Mind Reasoning: Developing models that infer human mental states from multimodal behavioral data. Embodied Assistance: Creating agents capable of assisting humans in physical environments through verbal and non-verbal collaboration. Learning from Human Feedback: Extracting reward-relevant preferences from rich human input to guide agent behavior. Social Scene Understanding: Recognizing and reasoning about group activities and social roles from visual and physical cues. Computational Social Cognition: Modeling how humans perceive and interpret social and physical interactions. Scientific Awards: Outstanding Paper Award at ACL 2024 for "MMToM-QA: Multimodal Theory of Mind Question Answering" Grants & Collaborations: Tianmin Shu has led or co-organized several high-impact workshops and tutorials, including the NeurIPS 2023 Tutorial on "Language Models Meet World Models" and the RSS 2024 Workshop on "Social Intelligence in Humans and Robots". His lab has also developed open-source platforms like VirtualHome-Social and SimWorld for multi-agent interaction research. Lab & Team: As director of the Social Cognitive AI (SCAI) Lab at Johns Hopkins University, Tianmin Shu leads a multidisciplinary team focused on building socially intelligent systems. His lab is located in Malone Hall 213 and collaborates closely with the Departments of Computer Science and Cognitive Science.
