If you made any changes in Pure, your changes will be visible here soon.

Fingerprint Dive into the research topics where Jason Dean Lee is active. These topic labels come from the works of this person. Together they form a unique fingerprint.

  • 1 Similar Profiles
Neural networks Engineering & Materials Science
Lasso Mathematics
Factorization Engineering & Materials Science
Chemical activation Engineering & Materials Science
Random Projection Mathematics
Polynomials Engineering & Materials Science
Sketching Mathematics
Structure Learning Mathematics

Network Recent external collaboration on country level. Dive into details by clicking on the dots.

Projects 2019 2023

Research Output 2010 2020

  • 850 Citations
  • 14 h-Index
  • 14 Conference article
  • 12 Article
  • 8 Conference contribution
  • 6 Paper
1 Citation (Scopus)

Stochastic Subgradient Method Converges on Tame Functions

Lee, J. D., Feb 1 2020, In : Foundations of Computational Mathematics. 20, 1, p. 119-154 36 p.

Princeton University

Research output: Contribution to journalArticle

Subgradient Method
Stochastic Methods
Locally Lipschitz Function
6 Citations (Scopus)

Communication-Efficient Distributed Statistical Inference

Lee, J. D., Apr 3 2019, In : Journal of the American Statistical Association. 114, 526, p. 668-681 14 p.

Research output: Contribution to journalArticle

Statistical Inference
Bayesian inference
Markov Chain Monte Carlo Algorithms
5 Citations (Scopus)

First-order methods almost always avoid strict saddle points

Lee, J. D., Jul 1 2019, In : Mathematical Programming. 176, 1-2, p. 311-337 27 p.

Research output: Contribution to journalArticle

Dynamical systems
Coordinate Descent
6 Citations (Scopus)

Theoretical insights into the optimization landscape of over-parameterized shallow neural networks

Lee, J. D., Feb 2019, In : IEEE Transactions on Information Theory. 65, 2, p. 742-769 28 p., 8409482.

Research output: Contribution to journalArticle

neural network
Neural networks
Chemical activation
Gaussian distribution

Adding one neuron can eliminate all bad local minima

Lee, J. D., Jan 1 2018, In : Advances in Neural Information Processing Systems. 2018-December, p. 4350-4360 11 p.

Research output: Contribution to journalConference article

Neural networks