Sotabase

Career

· Associate Professor, Princeton University2019–
· PhD in Computer Science, UC Berkeley2013–2019

Publications (118)

How to Escape Saddle Points Efficiently
International Conference on Machine Learning · 2017
903
cited
Provably Efficient Reinforcement Learning with Linear Function Approximation
Annual Conference Computational Learning Theory · 2019
629
cited
Intrinsic disc emission and the soft X‐ray excess in active galactic nuclei
2011
376
cited
Gradient Descent Can Take Exponential Time to Escape Saddle Points
Neural Information Processing Systems · 2017
253
cited
On the Theory of Transfer Learning: The Importance of Task Diversity
Neural Information Processing Systems · 2020
242
cited
Provable Meta-Learning of Linear Representations
International Conference on Machine Learning · 2020
211
cited
Sampling can be faster than optimization
Proceedings of the National Academy of Sciences of the United States of America · 2018
196
cited
On Nonconvex Optimization for Machine Learning
Journal of the ACM · 2019
191
cited
Stochastic Cubic Regularization for Fast Nonconvex Optimization
Neural Information Processing Systems · 2017
180
cited
A Short Note on Concentration Inequalities for Random Vectors with SubGaussian Norm
arXiv.org · 2019
160
cited
Local Maxima in the Likelihood of Gaussian Mixture Models: Structural Results and Algorithmic Consequences
Neural Information Processing Systems · 2016
143
cited
A Combined Optical and X-ray Study of Unobscured Type 1 AGN. I. Optical Spectra and SED Modeling
2011
140
cited
Streaming PCA: Matching Matrix Bernstein and Near-Optimal Finite Sample Guarantees for Oja's Algorithm
Annual Conference Computational Learning Theory · 2016
133
cited
A combined Optical and X-ray Spectra Study for Type 1 AGN. III. Broadband SED Properties
2012
121
cited
Stability and Convergence Trade-off of Iterative Optimization Algorithms
arXiv.org · 2018
100
cited
Provable Efficient Online Matrix Completion via Non-convex Stochastic Gradient Descent
Neural Information Processing Systems · 2016
85
cited
A powerful flare from Sgr A* confirms the synchrotron nature of the X-ray emission
2017
83
cited
Efficient Algorithms for Large-scale Generalized Eigenvector Computation and Canonical Correlation Analysis
International Conference on Machine Learning · 2016
82
cited
Faster Eigenvector Computation via Shift-and-Invert Preconditioning
International Conference on Machine Learning · 2016
80
cited
Sample-Efficient Reinforcement Learning of Undercomplete POMDPs
Neural Information Processing Systems · 2020
75
cited
Sotabase
Chi Jin | Researcher Profile | Sotabase | Sotabase