Will Grathwohl

Computer Science · Physical Sciences

14h-index1.3kcitations28works0.02yr avg
OpenAlex↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Generative Adversarial Networks and Image Synthesis(12), Topic Modeling(6), Neural Networks and Applications(4), Anomaly Detection Techniques and Applications(4), Adversarial Robustness in Machine Learning(4)

Publications28 total

A Fresh Take on Stale Embeddings: Improving Dense Retriever Training with Corrector Networks
arXiv (Cornell University)·2024Open Access
Variance reduction of diffusion model's gradients with Taylor approximation-based control variate
arXiv (Cornell University)·2024Open Access
Denoising Diffusion Samplers
arXiv (Cornell University)·2023· 3 citedOpen Access
Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMC
arXiv (Cornell University)·2023· 14 citedOpen Access
Continuous diffusion for categorical data
arXiv (Cornell University)·2022· 21 citedOpen Access
Self-conditioned Embedding Diffusion for Text Generation
arXiv (Cornell University)·2022· 21 citedOpen Access
Learning to Navigate Wikipedia by Taking Random Walks
arXiv (Cornell University)·2022· 1 citedOpen Access
Score-Based Diffusion meets Annealed Importance Sampling
arXiv (Cornell University)·2022· 3 citedOpen Access
Optimal Design of Stochastic DNA Synthesis Protocols based on Generative Sequence Models
2021· 15 citedOpen Access
Directly Training Joint Energy-Based Models for Conditional Synthesis and Calibrated Prediction of Multi-Attribute Data
arXiv (Cornell University)·2021Open Access
Oops I Took A Gradient: Scalable Sampling for Discrete Distributions
arXiv (Cornell University)·2021· 8 citedOpen Access
Learning the Stein Discrepancy for Training and Evaluating Energy-Based Models without Sampling
International Conference on Machine Learning·2020· 20 cited
No MCMC for me: Amortized sampling for fast and stable training of energy-based models
arXiv (Cornell University)·2020· 7 citedOpen Access
Learning the Stein Discrepancy for Training and Evaluating Energy-Based Models without Sampling
arXiv (Cornell University)·2020· 15 citedOpen Access
Cutting out the Middle-Man: Training and Evaluating Energy-Based Models without Sampling
arXiv (Cornell University)·2020· 9 citedOpen Access
Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One
arXiv (Cornell University)·2019· 86 citedOpen Access
Understanding the Limitations of Conditional Generative Models
arXiv (Cornell University)·2019· 13 citedOpen Access
Invertible Residual Networks
International Conference on Machine Learning·2019· 192 cited
Deep Reinforcement Learning and Simulation as a Path Toward Precision Medicine
Journal of Computational Biology·2019· 60 citedOpen Access
Invertible Residual Networks
arXiv (Cornell University)·2018· 170 citedOpen Access
page 1 of 2Next →

Frequent Co-authors

David Duvenaud(11), Richard S. Zemel(6), Jörn-Henrik Jacobsen(5), Ricky T. Q. Chen(4), Kuan-Chieh Wang(4), Randal Douc(4), Kevin Swersky(3), Brenden K. Petersen(3), Jiachen Yang(3), Chase Cockrell(3), Cláudio Santiago(3), Gary An(3), Daniel Faissol(3), Sander Dieleman(3), Robin Strudel(3), Rob Fergus(3), Jesse Bettencourt(2), Ilya Sutskever(2), Jens Behrmann(2), Joern-Henrik Jacobsen(2)