جست‌وجوی استاد و پژوهشگر

Anca Draganمشاهده پروفایل

دانشیار

• Human-AI Interaction
• Human-Robot Interaction
• AI Safety and Alignment
• +۵ مورد دیگر

Anca Dragan is an Associate Professor in the Department of Electrical Engineering and Computer Sciences at the University of California, Berkeley, where she runs the InterACT Lab focused on algorithms for human-AI and human-robot interaction. Currently on leave from Berkeley, she leads AI Safety and Alignment at Google DeepMind, overseeing safety for Gemini models and preparing for future advancements. Dragan has been a co-PI of the Center for Human-Compatible AI and served on the steering committee for the Berkeley AI Research (BAIR) Lab. B.Sc. in Computer Science from Jacobs University Bremen, Germany Ph.D. from Carnegie Mellon University Dr. Dragan's research focuses on enabling AI agents to work effectively with, around, and in support of people. Her work bridges robotics, machine learning, and game theory to create systems that better understand human preferences and coordinate with users. Key areas include AI alignment (ensuring AI does what people actually want), learning reward functions from diverse human feedback forms, and developing algorithms for human-AI collaboration across domains like autonomous vehicles, brain-machine interfaces, and recommender systems. Her research emphasizes maintaining uncertainty about human preferences and accounting for the plurality of human values. Dr. Dragan's recent publications reveal a strong focus on addressing fundamental challenges in AI safety and alignment. Her work spans theoretical foundations of reward learning, practical implementations for human-AI coordination, and critical examinations of limitations in current approaches. There's a clear trajectory toward more robust, safe, and value-aligned AI systems that can handle complex human preferences while avoiding both present-day harms and potential catastrophic risks. IEEE RAS Early Academic Career Award in Robotics and Automation (2021) McEntyre Award for Excellence in Teaching (2020) PECASE (Presidential Early Career Award for Science and Engineering) (2019) Sloan Fellowship (2018) NSF CAREER Award (2017) Okawa Foundation Award (2017) MIT Tech Review 35 Innovators Under 35 (2017) Multiple best paper awards at top robotics and AI conferences Dr. Dragan has mentored numerous successful students who have gone on to faculty positions at MIT, Stanford, CMU, and Princeton, as well as industry roles at DeepMind, Waymo, and Meta. Her advising philosophy emphasizes both technical rigor and consideration of broader societal impacts. She has secured significant research funding including NSF CAREER, ONR Young Investigator, and Okawa Foundation awards, supporting work on human-AI interaction and alignment. Dragan has also consulted for Waymo for six years, helping develop roadmaps for deploying increasingly learning-based safety-critical systems. Dr. Dragan leads the InterACT Lab at UC Berkeley, which has produced influential work on Cooperative Inverse Reinforcement Learning, Inverse Reward Design, and other foundational concepts in human-AI interaction. The lab's research has significantly shaped the field of AI alignment, with applications spanning autonomous vehicles that coordinate with human drivers, brain-machine interfaces that adapt to user needs, and language models that better understand human preferences. Current work focuses on scaling safety approaches as AI capabilities advance, ensuring alignment keeps pace with technological progress.

Benjamin Eysenbachمشاهده پروفایل

استاد

• Reinforcement Learning
• Self-Supervised Learning
• Machine Learning
• +۲ مورد دیگر

Benjamin Eysenbach leads the Princeton Reinforcement Learning Lab, where he designs algorithms that enable artificial intelligence systems to learn intelligent behaviors through trial-and-error, specializing in self-supervised methods that eliminate the need for human labels. He joined Princeton after completing his PhD in machine learning at Carnegie Mellon University under Ruslan Salakhutdinov and Sergey Levine, supported by the NSF Graduate Research Fellowship and Hertz Fellowship. His research bridges fundamental machine learning principles with practical applications in robotics and decision-making systems. Eysenbach's research focuses on developing self-supervised reinforcement learning algorithms that enable autonomous skill acquisition without external rewards. His investigations span contrastive learning methods, temporal abstraction techniques, and scalable architectures for goal-conditioned behaviors. These innovations aim to create more efficient and generalizable learning systems that can discover useful behaviors from unlabeled experience. Eysenbach's publications demonstrate consistent advancement in self-supervised RL methodologies, with recent work focusing increasingly on temporal abstraction and representation learning theory. His research shows progression from foundational contrastive RL frameworks toward more sophisticated analyses of generalization properties and uncertainty quantification. The 2025 works indicate expanding investigation into hierarchical control, probabilistic alignment, and hyper-deep network architectures. Eysenbach has been recognized with prestigious awards including the Hertz Fellowship and NSF Graduate Research Fellowship, supporting his doctoral research in self-supervised RL methodologies. His work has been presented at top machine learning conferences including NeurIPS, ICML, and ICLR. As director of the Princeton Reinforcement Learning Lab, Eysenbach oversees research initiatives in self-supervised RL, including projects on intention-conditioned modeling, horizon generalization, and contrastive learning frameworks. He has secured funding from the Princeton AI Lab to study neural correlates of temporal contrast in decision-making. Eysenbach teaches courses in reinforcement learning and has developed new benchmarks like JaxGCRL to accelerate research in goal-conditioned RL.

Feras Saadمشاهده پروفایل

استادیار

• Probabilistic Programming
• Artificial Intelligence
• Programming Languages
• +۴ مورد دیگر

Feras Saad is an Assistant Professor in the Computer Science Department at Carnegie Mellon University, affiliated with the Principles of Programming and Artificial Intelligence groups. He received his Ph.D. in Computer Science from the Massachusetts Institute of Technology (MIT) in 2022, where his dissertations on probabilistic programming systems earned him the George M. Sprowls PhD Thesis Award and Charles & Jennifer Johnson MEng Thesis Award. His research focuses on developing scalable computing systems for probabilistic modeling and inference, integrating ideas from programming languages and probabilistic AI. Key research themes include probabilistic programming languages, automated probabilistic model discovery, statistical estimation and testing, random sampling algorithms, and applications in science and engineering. His lab explores new techniques to improve reasoning systems through automation, accuracy, and scale. Dr. Saad has published extensively in top venues including PLDI, POPL, ICML, and Nature Communications. His work spans foundational computational questions to practical software systems for probabilistic inference. Research trends show consistent focus on bridging theoretical computer science with practical applications in probabilistic modeling, with recent advances in random sampling algorithms and probabilistic programming systems. Awards and honors include: George M. Sprowls PhD Thesis Award in Artificial Intelligence and Decision Making (2023) Charles & Jennifer Johnson MEng Thesis Award in Computer Science (2017) Editor's Highlight for Nature Communications paper (2024) He currently advises graduate students Gaurav Arya and Thomas Draper in the Probabilistic Computing Systems Lab. His research is supported by software libraries including GenSQL, BayesNF, and SPPL that enable practical applications across scientific domains.

Ron Yurkoمشاهده پروفایل

عضو هیئت علمی

• Sports Analytics
• Natural Language Processing & Large Language Models
• Biostatistics & Epidemiology
• +۴ مورد دیگر

Ron Yurko is an Assistant Teaching Professor in the Department of Statistics & Data Science at Carnegie Mellon University , where he focuses on methods at the intersection of statistical inference and machine learning. He serves as Director of the Carnegie Mellon Sports Analytics Center (CMSAC) , overseeing initiatives like the sports analytics research lab, CMSACamp , and the annual Sports Analytics Conference. Research Interests His work bridges Sports Analytics and Natural Language Processing , with applications in Biostatistics , Public Policy , and Statistical Pedagogy . Recent projects include evaluating NFL player positioning ( NFL Ghosts ), analyzing language patterns in LLM outputs, and developing fractional tackle metrics for defensive evaluation. Google Scholar articles reveal deeper methodological contributions to Bayesian modeling for directional movement analysis, multilevel models for snap timing, and spatio-temporal frameworks in football event modeling. Academic Background PhD in Statistics (2022) from Carnegie Mellon University BS in Statistics with University Honors (2015) from Carnegie Mellon Additional Contributions He is developing a textbook Statistical Methods in Sports Analytics (2027), maintains the Statistical Thinking in Sports Analytics newsletter, and teaches course 36-460/660 at CMU.

Liu Yangمشاهده پروفایل

پژوهشگر ارشد

• Theoretical Machine Learning
• Theoretical Computer Science
• Statistical Learning Theory
• +۵ مورد دیگر

Liu Yang is a postdoctoral fellow in the Computer Science Department at Carnegie Mellon University , with a PhD from CMU under Avrim Blum and Jaime Carbonell. His research focuses on Theoretical Machine Learning and Theoretical Computer Science , exploring areas like Statistical Learning Theory , Property Testing , and Algorithmic Economics . He has contributed to active learning , transfer learning , and online pricing problems through mathematical frameworks. Research Interests : Liu's work bridges Computational Learning Theory with Algorithmic Economics , including: Mathematical theories for active property testing of Boolean functions Transfer learning with applications to online allocation and pricing Analysis of convex losses and statistical identifiability in learning Developing Buys-in-Bulk models for active learning efficiency Service and Teaching : He has served on program committees for ICML 2012-2013 , reviewed for top-tier venues, and taught courses like Graduate Algorithms and Modern Computer Algebra at CMU. He also co-developed the DistLearnKit MATLAB toolkit for distance metric learning.

Jiashun Jinمشاهده پروفایل

استاد

• Statistics
• Big Data Analytics
• Network Analysis
• +۴ مورد دیگر

Jiashun Jin is a Professor of Statistics and Affiliated Professor of Machine Learning at Carnegie Mellon University, where he leads research in statistical inference for Big Data with a focus on 'Rare and Weak' signal regimes. He received his Ph.D. in Statistics from Stanford University in 2003. His research interests span: Development of novel methods like Higher Criticism, Graphlet Screening, IF-PCA, and SCORE for high-dimensional data Applications in genomics (cancer classification, SNP analysis), cosmology (non-Gaussian detection in CMB), and network science Theoretical frameworks including phase diagrams to characterize statistical detectability boundaries Recent publications (2021-2023) demonstrate strong emphasis on network community detection, text analysis, genetic association studies, and COVID-19 resource allocation, with consistent focus on statistical optimality in high-dimensional settings. Honors and awards include: NSF CAREER Award (2007) IMS Tweedie Award (2009) IMS Fellow (2011) ASA Fellow (2019) ICCM Distinguished Paper Award (2020) He has developed a significant dataset of 83,381 statistical publications for coauthorship/citation network analysis and has delivered invited lectures including the IMS Medallion Lecture (2015).

Andrej Risteskiمشاهده پروفایل

دانشیار

• Machine Learning
• Statistics
• Theoretical Computer Science
• +۵ مورد دیگر

Andrej Risteski is an Associate Professor at the Machine Learning Department of Carnegie Mellon University (CMU) since 2025. He previously held the Norbert Wiener Research Fellow position jointly between the Applied Math Department and IDSS at MIT (2017–2019) after completing his PhD in Computer Science at Princeton University (2012–2017) under Sanjeev Arora . His research focuses on the intersection of machine learning, statistics, and theoretical computer science , emphasizing generative models, representation learning, and out-of-distribution generalization with applications to natural language processing and scientific domains . His recent publications explore edge embeddings in Graph Neural Networks (GNNs) , score matching efficiency , and theoretical foundations of diffusion models . Key contributions include analyzing computational bottlene.com/activities/statistical-and-computational-challenges-in-probabilistic-scientific-machine-learning-sciml/">NSF CAREER Award , DOE Computational Science Graduate Fellowship for Stephen Huan, and co-organizing the COLT workshop on Theory of AI for Scientific Computing . He advises PhD students across Machine Learning, Computer Science, and Mathematics , including Bingbin Liu (Kempner Institute Fellow) and Elan Rosenfeld (Google Research Scientist). Teaching includes Probabilistic Graphical Models and Advanced Deep Learning at CMU, plus Applied Mathematics at MIT. His work is supported by NSF, DoD , and OpenAI Superalignment grants. Education : PhD in Computer Science (Princeton), BSc in Computer Science (Princeton) Current Positions : Associate Professor, CMU Machine Learning Department Former Positions : Norbert Wiener Fellow, MIT IDSS & Applied Mathematics Research Areas : Generative Models (GANs, Diffusion Models) Representation Learning Out-of-Distribution Generalization Neural Language Models AI for Scientific Applications Sampling and Optimization Algorithms Scientific Awards : NSF CAREER Award (2023) Google Research Award (2024) Amazon Research Award (2022) OpenAI Superalignment Grant (2023) Recent Talks (2023–2025): "Architectural Nuances and Benchmark Gaps in Scientific ML" (UC Berkeley, 2025) "The Statistical Cost of Score-Based Losses" (Simons Institute Boot Camp, 2024) "Neural Networks for PDEs" (ETH Zurich, 2024) "Discernible Patterns in Transformers" (Theory of Interpretable AI, 2024) He leads a research group producing work at the interface of computational complexity and graph learning , with empirical validation on topological bottlenecks and hub node dynamics . Current projects include ICML 2025 paper on edge embeddings in GNNs and COLT 2025 workshop on AI for Scientific Computing co-organized with MIT, Duke, and ETH Zurich collaborators.

Cosma Shaliziمشاهده پروفایل

دانشیار

• Nonparametric prediction of time series
• Learning theory
• Nonlinear dynamics
• +۱۵ مورد دیگر

Cosma Shalizi is an Associate Professor in the Statistics Department and Machine Learning Department at Carnegie Mellon University, and an External Professor at the Santa Fe Institute. His work bridges statistics, machine learning, and complex systems theory, with applications spanning neuroscience, statistical mechanics, and social networks. Shalizi's research focuses on nonparametric prediction of time series, learning theory, information theory, and causal inference. He has made significant contributions to computational mechanics, developing algorithms like CSSR (Causal State Splitting Reconstruction) for identifying optimal predictive states in complex systems. His work extends to heavy-tailed distributions, network analysis, and the statistical foundations of complex systems. He has pioneered methods for quantifying self-organization and developing nonparametric approaches to spatio-temporal prediction. His recent publications reveal a trend toward increasingly interdisciplinary work, connecting network science with causal inference, statistical learning theory with macroeconomic forecasting, and information theory with biomedical applications. His work consistently emphasizes rigorous statistical methodology applied to complex, dependent data structures across diverse scientific domains. Winner of the Best Student Paper and Best Poster awards Shalizi has advised students including Georg Goerg, who extended spatio-temporal prediction techniques to continuous-valued fields, and George Montañez, who developed fast approximate algorithms for prediction and explored information-theoretic explanations for machine learning. His collaborative network spans statistics, physics, neuroscience, and social sciences, reflecting his interdisciplinary approach to complex systems. His work with collaborators has led to significant contributions in network analysis, causal inference in social networks, and the development of nonparametric methods for complex data structures. He maintains active research programs in statistical network modeling, time series analysis, and the application of information-theoretic approaches to diverse scientific problems.

Fabio Cozmanمشاهده پروفایل

استاد

• Artificial Intelligence
• Machine Learning
• Probabilistic Reasoning
• +۷ مورد دیگر

Fabio Cozman is a Full Professor at the School of Engineering, University of São Paulo (USP), and Director of the Center for Artificial Intelligence at USP. He holds a PhD from Carnegie Mellon University (1996) and completed his engineering degree at Poli-USP. His research spans artificial intelligence, machine learning, and probabilistic reasoning, with significant contributions to knowledge representation under uncertainty, credal networks, and physics-informed AI. Current research focuses on AI interpretability , AI and societal impact , and ocean hazard prediction . Key publications cover probabilistic logic programming , credal networks , and Bayesian network extensions . He is a founding member and former president of the Society for Imprecise Probability Theory and Applications (SIPTA) and has served as Associate Editor for multiple AI journals. His work integrates theoretical advancements in probabilistic modeling with practical applications in robotics, computer vision, and environmental systems, particularly focused on the Brazilian maritime territory through projects like the Blue Amazon Brain (BLAB) architecture.

Jan Hoffmannمشاهده پروفایل

دانشیار

• Programming Languages
• Formal Methods
• Security and Privacy

Jan Hoffmann is an Associate Professor in the Computer Science Department at Carnegie Mellon University. His research develops quantitative program analysis techniques to ensure software reliability, efficiency, and security. Key focus areas include resource-aware programming languages, probabilistic program verification, and scalable static analysis methods for modern computing environments. Dr. Hoffmann leads the Carnegie Mellon Resource Analysis Group, advising PhD students on projects spanning energy-aware computing, secure compilation, and probabilistic inference. His work bridges formal methods with practical system design, emphasizing compositional verification and automated reasoning. Recent publications advance CUDA kernel optimization, programmable Markov chain Monte Carlo methods, and robust resource bound synthesis. This research establishes foundations for predictable performance in concurrent systems and AI-driven applications.

• Robust Bayesian Inference
• Statistical Network Analysis
• Hierarchical Bayesian Analysis
• +۲ مورد دیگر

Neil Spencer is an Assistant Professor in the Department of Statistics at the University of Connecticut. His primary research focuses on robust Bayesian inference, statistical network analysis, hierarchical Bayesian modeling, and efficient Bayesian computation, with applications in forensic footwear analysis and network data. He teaches statistical computing courses including STAT5410 for the Master of Data Science program. His educational background includes a joint PhD in Statistics and Machine Learning from Carnegie Mellon University, an MSc in Statistics from the University of British Columbia, and a BScH in Mathematics and Statistics from Acadia University. Previously, he was a postdoctoral researcher in Biostatistics at Harvard School of Public Health. Dr. Spencer's recent publications demonstrate a strong focus on Bayesian methodologies applied to network modeling, computational statistics, and experimental design. His work frequently appears in top-tier statistics journals such as the Annals of Statistics.

Aaditya Ramdasمشاهده پروفایل

دانشیار

• Statistics
• Machine Learning
• Post-selection inference
• +۹ مورد دیگر

Aaditya Ramdas is an Associate Professor in the Department of Statistics and Data Science and the Machine Learning Department at Carnegie Mellon University. He is also a visiting academic at Amazon Research. His work bridges theoretical statistics and practical machine learning, focusing on algorithms with strong guarantees for post-selection inference, game-theoretic statistics, and predictive uncertainty quantification. Applied domains include privacy, neuroscience, genetics, and auditing. He has received prestigious awards such as the Sloan Fellowship, IMS Peter Gavin Hall Prize, and NSF CAREER Award. His research spans foundational areas like sequential testing, multiple hypothesis testing, and conformal prediction, with recent work emphasizing anytime-valid inference and robust statistical methods. Affiliations: Dietrich College of Humanities and Social Sciences, Amazon Research Labs/Teams: DELPHI Lab Group (specializing in public health forecasting), CSAFE Lab Group (computational statistics). Key contributions include advancing e-value based methods for hypothesis testing and developing scalable techniques for uncertainty quantification in machine learning models. His work often intersects interdisciplinary fields, such as auditing AI systems for fairness and privacy compliance.

Zach Bransonمشاهده پروفایل

عضو هیئت علمی

• Experimental Design
• Causal Inference
• Nonparametric Methods
• +۸ مورد دیگر

Zach Branson is an Associate Teaching Professor and Assistant Director for the Undergraduate Program at Carnegie Mellon University's Department of Statistics & Data Science within the Dietrich College of Humanities and Social Sciences. His work bridges theoretical and applied statistics, with a focus on experimental design and causal inference. PhD in Statistics , Harvard University (2019) BS in Economics and Statistics , Carnegie Mellon University (2014) BA in Professional Writing , Carnegie Mellon University (2014) Branson's research centers on experimental design and causal inference, particularly addressing questions like "Does a treatment cause a change in outcomes?" His methodological interests include covariate balance , matching , randonization tests , and regression discontinuity designs . Applications span education , epidemiology , mental health , and text analysis . His recent publications emphasize causal inference in educational and social sciences, with methodological innovations in rerandomization, propensity score trimming, and regression discontinuity designs. Key themes include improving covariate balance, handling continuous treatments, and applying Bayesian nonparametric approaches to spatial data. NSF Graduate Research Fellowship (for PhD work on experimental design) Branson actively engages in teaching causality and statistical communication to undergraduates, advocating for capstone courses (e.g., 36-490, 36-493, 36-497) to foster practical data science skills. He also contributed to public-facing statistical pedagogy through "data science portfolio" projects.

Larry Wassermanمشاهده پروفایل

استاد

• nonparametric inference
• causal inference
• machine learning
• +۶ مورد دیگر

Larry Wasserman is a UPMC University Professor at Carnegie Mellon University, jointly appointed in the Department of Statistics and Data Science and the Machine Learning Department. He received his Ph.D. from the University of Toronto in 1988 and is recognized as one of the leading statisticians of his generation. His research spans theoretical and applied statistics, with core interests in: Foundational inference : Nonparametric methods, asymptotic theory, causal frameworks Modern applications : Machine learning, high-dimensional statistics, astrostatistics Interdisciplinary domains : Bioinformatics, genomics, physical sciences via the STAMPS group His recent publications demonstrate strong emphasis on causal methodology, optimal transport, and robust inference, with applications ranging from particle physics to genomic analysis. Articles frequently develop novel nonparametric techniques with minimax optimality guarantees. Award highlights include: COPSS Presidents' Award (1999) - Top honor for statisticians under 40 CRM-SSC Prize (2002) - Landmark contributions to statistics Fellowships: American Statistical Association, Institute of Mathematical Statistics, AAAS He leads the Statistical Machine Learning Theory Group and founded STAMPS (Statistical Methods for Physical Sciences). His textbooks All of Statistics and All of Nonparametric Statistics are widely used in graduate programs globally.

Anderson Ye Zhangمشاهده پروفایل

استادیار

• Spectral Analysis
• Group Synchronization
• Ranking from Pairwise Comparisons
• +۴ مورد دیگر

Anderson Ye Zhang is an Assistant Professor in the Department of Statistics and Data Science at the Wharton School, University of Pennsylvania, with a secondary appointment in the Department of Computer and Information Science. He holds a PhD from Yale University and previously served as a William H. Kruskal Instructor at the University of Chicago. His research focuses on the theoretical and applied aspects of statistics and machine learning, emphasizing spectral methods, synchronization problems, clustering, and network analysis. Education: PhD in Statistics and Data Science, Yale University (2018) Bachelor's degree from Zhejiang University (2012) Research Interests: His work addresses foundational challenges in spectral analysis, group synchronization (e.g., phase synchronization, permutation synchronization), ranking systems, and high-dimensional clustering. He develops algorithms with provable guarantees for problems such as Gaussian mixture models, stochastic block models, and item response theory. Recent Trends in Publications (2021–2025): His articles emphasize spectral methods for synchronization and clustering, with a focus on optimality in high-noise regimes, privacy-preserving learning, and efficient algorithms for complex data structures like human response models and anisotropic covariances. His work bridges statistical theory and computational practice. Awards & Grants: 2025: Sloan Research Fellowship, NSF CAREER Award 2019: New Researcher Award (ICSA) 2018: Francis J. Anscombe Award Grants: NSF DMS-2440180 (CAREER: Statistical Inference in Group Actions) NSF DMS-2112988 (Ranking from Comparisons) Academic Contributions: He has advised multiple collaborative research projects with postdocs and students, particularly on spectral methods and synchronization problems. His teaching spans advanced statistical theory and linear models at Wharton and the University of Chicago.