Philip M. Long

Google (United States) — United States

Computer Science · Physical Sciences

39h-index7.3kcitations180works0.02yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

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

Research Topics

Machine Learning and Algorithms(102), Algorithms and Data Compression(35), Machine Learning and Data Classification(27), Stochastic Gradient Optimization Techniques(26), Complexity and Algorithms in Graphs(18)

Publications180 total

Corrigendum to “Prediction, learning, uniform convergence, and scale-sensitive dimensions” [J. Comput. Syst. Sci. 56 (2) (1998) 174–190]
Journal of Computer and System Sciences·2023
Sharpness-Aware Minimization and the Edge of Stability
arXiv (Cornell University)·2023Open Access
The Dynamics of Sharpness-Aware Minimization: Bouncing Across Ravines and Drifting Towards Wide Minima
arXiv (Cornell University)·2022· 2 citedOpen Access
Deep Linear Networks can Benignly Overfit when Shallow Ones Do
arXiv (Cornell University)·2022· 2 citedOpen Access
The Perils of Being Unhinged: On the Accuracy of Classifiers Minimizing a Noise-Robust Convex Loss
Neural Computation·2022Open Access
Oracle lower bounds for stochastic gradient sampling algorithms
Bernoulli·2022· 6 cited
The perils of being unhinged: On the accuracy of classifiers minimizing\n a noise-robust convex loss
arXiv (Cornell University)·2021Open Access
Foolish Crowds Support Benign Overfitting
arXiv (Cornell University)·2021· 1 citedOpen Access
The Interplay Between Implicit Bias and Benign Overfitting in Two-Layer\n Linear Networks
arXiv (Cornell University)·2021· 2 citedOpen Access
The Interplay Between Implicit Bias and Benign Overfitting in Two-Layer Linear Networks
arXiv (Cornell University)·2021Open Access
When does gradient descent with logistic loss interpolate using deep networks with smoothed ReLU activations
Conference on Learning Theory·2021· 1 cited
Properties of the After Kernel
arXiv (Cornell University)·2021· 1 citedOpen Access
When does gradient descent with logistic loss interpolate using deep networks with smoothed ReLU activations?
arXiv (Cornell University)·2021· 5 citedOpen Access
Collider bias in COVID-19 studies: Explaining the smoking paradox
Pharmacoepidemiology and Drug Safety·2021
Superlinear Integrality Gaps for the Minimum Majority Problem
SIAM Journal on Discrete Mathematics·2021
Failures of Model-dependent Generalization Bounds for Least-norm Interpolation
Journal of Machine Learning Research·2021· 1 cited
When does gradient descent with logistic loss find interpolating two-layer networks?
arXiv (Cornell University)·2020· 4 citedOpen Access
Failures of model-dependent generalization bounds for least-norm interpolation
arXiv (Cornell University)·2020· 15 citedOpen Access
Finite-sample Analysis of Interpolating Linear Classifiers in the Overparameterized Regime
arXiv (Cornell University)·2020· 22 citedOpen Access
Finite-sample Analysis of Interpolating Linear Classifiers in the Overparameterized Regime
Journal of Machine Learning Research·2020· 31 cited
page 1 of 9Next →

Frequent Co-authors

Edison T. Liu(6), Rocco A. Servedio(6), Vinsensius B. Vega(5), Peter L. Bartlett(5), Manfred K. Warmuth(4), David P. Helmbold(4), Nicolò Cesa‐Bianchi(3), Hanie Sedghi(3), N. Littlestone(3), Lisa M. McShane(2), Yijun Ruan(2), Chia Lin Wei(2), Ai Ee Ling(2), Tao Zhang(2), Ee Chee Ren(2), Lawrence W. Stanton(2), Lance D. Miller(2), Amirul Islam(2), Aravind Srinivasan(2), Robert C. Williamson(2)