World's Best Scientists 2026 revealed!
Jascha Sohl-Dickstein

Jascha Sohl-Dickstein

D-Index & Metrics

Computer Science

D-Index
56
Citations
15901
World Ranking
4003
National Ranking
1907

Overview

Jascha Sohl-Dickstein is affiliated with Google in the United States. Their research spans primarily the field of Computer Science, with a focus on subfields including Artificial Intelligence, Computer Vision and Pattern Recognition, Statistical and Nonlinear Physics, Signal Processing, and Computational Theory and Mathematics.

The scientist's work addresses several key topics such as Neural Networks and Applications, Machine Learning and Data Classification, Advanced Neural Network Applications, Domain Adaptation and Few-Shot Learning, Topic Modeling, Natural Language Processing Techniques, and Stochastic Gradient Optimization Techniques.

Jascha Sohl-Dickstein has contributed extensively to academic literature, with notable recent papers including:

  • "Score-Based Generative Modeling through Stochastic Differential Equations," 2020, arXiv (Cornell University)
  • "Rethink reporting of evaluation results in AI," 2023, Science
  • "The large learning rate phase of deep learning: the catapult mechanism," 2020, arXiv (Cornell University)
  • "Levels of AGI for Operationalizing Progress on the Path to AGI," 2023, arXiv (Cornell University)
  • "Finite Versus Infinite Neural Networks: an Empirical Study," 2020, arXiv (Cornell University)

The most frequent publication venues for Sohl-Dickstein include arXiv (Cornell University), with 45 publications, Nature Communications, Science, Northern European Journal of Language Technology, and the Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence.

Jascha Sohl-Dickstein has collaborated on numerous research projects with co-authors such as Luke Metz, Roman Novak, Jaehoon Lee, Jeffrey Pennington, and Samuel S. Schoenholz. The number of collaborations ranges from 8 to 15 papers per co-author.

Best Publications

  • Density estimation using Real NVP

    Laurent Dinh;Jascha Sohl-Dickstein;Samy Bengio

  • Deep Unsupervised Learning using Nonequilibrium Thermodynamics

    Jascha Sohl-Dickstein;Eric Weiss;Niru Maheswaranathan;Surya Ganguli

  • Score-Based Generative Modeling through Stochastic Differential Equations

    Yang Song;Jascha Sohl-Dickstein;Diederik P Kingma;Abhishek Kumar

  • Deep knowledge tracing

    Chris Piech;Jonathan Bassen;Jonathan Huang;Surya Ganguli

  • Unrolled Generative Adversarial Networks

    Luke Metz;Ben Poole;David Pfau;Jascha Sohl-Dickstein

  • Wide Neural Networks of Any Depth Evolve as Linear Models Under Gradient Descent

    Jaehoon Lee;Lechao Xiao;Samuel S. Schoenholz;Yasaman Bahri

  • Deep Neural Networks as Gaussian Processes

    Jaehoon Lee;Yasaman Bahri;Roman Novak;Samuel S. Schoenholz

  • On the expressive power of deep neural networks

    Maithra Raghu;Ben Poole;Jon M. Kleinberg;Surya Ganguli

  • Exponential expressivity in deep neural networks through transient chaos

    Ben Poole;Subhaneil Lahiri;Maithreyi Raghu;Jascha Sohl-Dickstein

  • Mars Exploration Rover Athena Panoramic Camera (Pancam) investigation

    J.F. Bell;S. W. Squyres;Kenneth E. Herkenhoff;J.N. Maki

  • SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability

    Maithra Raghu;Justin Gilmer;Jason Yosinski;Jascha Sohl-Dickstein

  • Deep Knowledge Tracing

    Chris Piech;Jonathan Spencer;Jonathan Huang;Surya Ganguli

  • Sensitivity and Generalization in Neural Networks: an Empirical Study

    Roman Novak;Yasaman Bahri;Daniel A. Abolafia;Jeffrey Pennington

  • Adversarial Examples that Fool both Computer Vision and Time-Limited Humans

    Gamaleldin Fathy Elsayed;Shreya Shankar;Brian Cheung;Nicolas Papernot

  • Bayesian Deep Convolutional Networks with Many Channels are Gaussian Processes

    Roman Novak;Lechao Xiao;Jaehoon Lee;Yasaman Bahri

  • Statistical Mechanics of Deep Learning

    Yasaman Bahri;Jonathan Kadmon;Jeffrey Pennington;Sam S. Schoenholz

  • Dynamical Isometry and a Mean Field Theory of CNNs: How to Train 10,000-Layer Vanilla Convolutional Neural Networks

    Lechao Xiao;Yasaman Bahri;Jascha Sohl-Dickstein;Samuel S. Schoenholz

  • REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models

    George Tucker;Andriy Mnih;Chris J. Maddison;Dieterich Lawson

  • Measuring the Effects of Data Parallelism on Neural Network Training

    Christopher J. Shallue;Jaehoon Lee;Joseph M. Antognini;Jascha Sohl-Dickstein

  • Deep Information Propagation

    Samuel S. Schoenholz;Justin Gilmer;Surya Ganguli;Jascha Sohl-Dickstein

  • Neural Tangents: Fast and Easy Infinite Neural Networks in Python

    Roman Novak;Lechao Xiao;Jiri Hron;Jaehoon Lee

Frequent Co-Authors

Surya Ganguli
Surya Ganguli Stanford University
Ben Poole
Ben Poole Google (United States)
James F. Bell
James F. Bell Arizona State University
Bruno A. Olshausen
Bruno A. Olshausen University of California, Berkeley
Alexander G. Hayes
Alexander G. Hayes Cornell University
Jeffrey R. Johnson
Jeffrey R. Johnson Johns Hopkins University Applied Physics Laboratory
Ian Goodfellow
Ian Goodfellow Google (United States)
Walter Goetz
Walter Goetz University of Göttingen
Justin N. Maki
Justin N. Maki Jet Propulsion Lab

If you think any of the details on this page are incorrect, let us know.

Report an issue

We appreciate your kind effort to assist us to improve this page, it would be helpful providing us with as much detail as possible in the text box below:

Related Online Degrees & Career Pathways

Exploring online education can unlock a variety of career options for students interested in Computer Science. For those seeking a fast track into the workforce, there are easy certifications to get online which offer strong earning potential and can enhance your resume in a matter of months. These certifications cover areas like coding, cybersecurity, and cloud computing.

If you're looking to advance your qualifications quickly, consider quick masters degrees online. These accelerated programs allow busy professionals to gain advanced skills without putting their careers on hold. In fact, knowing which master's degree is most in demand in usa can help you target programs that lead to high-growth fields, such as artificial intelligence or data science.

For those not ready to commit to a four-year program, an associate degree online in Computer Science provides foundational skills that open doors to entry-level jobs and credits smoothly transfer to bachelor’s programs.

Best Scientists Citing Jascha Sohl-Dickstein

Trending Scientists