John Mitchell

About

John C. Mitchell is the Mary and Gordon Crary Family Professor at Stanford University, holding positions as a professor of computer science, with courtesy appointments in electrical engineering and education. He has previously served as Stanford Vice Provost for Teaching and Learning and as chair of the Computer Science Department. In his role as vice provost, he led initiatives involving over 500 faculty members and instructors on more than 1,000 online projects, and organized the Year of Learning to explore the future of teaching and learning at Stanford and beyond. Mitchell is a co-director of the Lytics Lab, Carta Lab, and Pathways Lab, where he works to improve educational outcomes through data-driven research and iterative design. His research focuses on programming languages, computer security and privacy, blockchain, machine learning, and technology for education. He has authored two textbooks and over 250 publications with more than 30,000 citations, and has served as a consultant, advisor, and editor-in-chief of the Journal of Computer Security. Mitchell's early work in online learning began in 2009 with the development of Stanford CourseWare, which supported interactive video and discussion, laying the foundation for Stanford's initial MOOCs and flipped classroom experiments.

Research topics

• Artificial Intelligence
• Computer Science
• Biology
• Genetics
• Machine Learning
• Bioinformatics
• Psychiatry
• Parallel computing
• Medicine
• Computational biology

Selected publications

• Genetic identification of cell types underlying brain complex traits yields insights into the etiology of Parkinson’s disease

  Nature Genetics · 2020 · 376 citations

  Genome-wide association studies have discovered hundreds of loci associated with complex brain disorders, but it remains unclear in which cell types these loci are active. Here we integrate genome-wide association study results with single-cell transcriptomic data from the entire mouse nervous system to systematically identify cell types underlying brain complex traits. We show that psychiatric disorders are predominantly associated with projecting excitatory and inhibitory neurons. Neurological…

  DOI

• Dissecting the Shared Genetic Architecture of Suicide Attempt, Psychiatric Disorders, and Known Risk Factors

  Biological Psychiatry · 2021 · 240 citations

  DOI

• Reinforcement Learning for the Adaptive Scheduling of Educational Activities

  2020 · 68 citations

  Senior authorCorresponding

  Adaptive instruction for online education can increase learning gains and decrease the work required of learners, instructors, and course designers. Reinforcement Learning (RL) is a promising tool for developing instructional policies, as RL models can learn complex relationships between course activities, learner actions, and educational outcomes. This paper demonstrates the first RL model to schedule educational activities in real time for a large online course through active learning. Our mod…

  PublisherDOI

• TrustLLM: Trustworthiness in Large Language Models

  arXiv (Cornell University) · 2024-01-10 · 52 citations

  preprintOpen access

  Large language models (LLMs), exemplified by ChatGPT, have gained considerable attention for their excellent natural language processing capabilities. Nonetheless, these LLMs present many challenges, particularly in the realm of trustworthiness. Therefore, ensuring the trustworthiness of LLMs emerges as an important topic. This paper introduces TrustLLM, a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, eval…

  PublisherOA PDFDOI

• Identifying and Mitigating the Security Risks of Generative AI

  Foundations and Trends® in Privacy and Security · 2023-12-14 · 43 citations

  articleOpen access

  Every major technical invention resurfaces the dual-use dilemma—the new technology has the potential to be used for good as well as for harm. Generative AI (GenAI) techniques, such as large language models (LLMs) and diffusion models, have shown remarkable capabilities (e.g., in-context learning, code-completion, and text-to-image generation and editing). However, GenAI can be used just as well by attackers to generate new attacks and increase the velocity and efficacy of existing attacks. This…

  PublisherDOI

Recent grants

• NIH Grant R01MH058820

  NIH · $2.3M · 2011

• NIH Grant R01MH059100

  NIH · $1.2M · 2005

• NIH Grant Z01SC006387

  NIH · $1.1M

Frequent coauthors

• Gerhard Weikum

  1715 shared

• Friedemann Mattern

  1715 shared

• David Hutchison

  Lancaster University

  1715 shared

• Bernhard Steffen

  TU Dortmund University

  1715 shared

• Moni Naor

  1715 shared

• Doug Tygar

  University of California, Berkeley

  1715 shared

• Demetri Terzopoulos

  1715 shared

• Oscar Nierstrasz

  1715 shared

Labs

• Lytics Lab, Carta Lab and Pathways LabPI

Education

• Ph.D., Computer Science

  Stanford University

• M.S., Computer Science

  Stanford University

• B.S., Computer Science

  Stanford University

Similar researchers at Stanford University

• Carl Wiemanh-113
• Rich Shavelsonh-82
• Philip Andrew Fisherh-80