Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Mengdi Wang

Mengdi Wang

Diffusion Approximations for Online Principal Component Estimation and Global Convergence

Aug 29, 2018
Chris Junchi Li, Mengdi Wang, Han Liu, Tong Zhang

Figure 1 for Diffusion Approximations for Online Principal Component Estimation and Global Convergence

Figure 2 for Diffusion Approximations for Online Principal Component Estimation and Global Convergence

Figure 3 for Diffusion Approximations for Online Principal Component Estimation and Global Convergence

* Appeared in NIPS 2017

Access Paper or Ask Questions

Spectral State Compression of Markov Processes

Aug 28, 2018
Anru Zhang, Mengdi Wang

Figure 1 for Spectral State Compression of Markov Processes

Figure 2 for Spectral State Compression of Markov Processes

Figure 3 for Spectral State Compression of Markov Processes

Figure 4 for Spectral State Compression of Markov Processes

Access Paper or Ask Questions

Variance Reduction Methods for Sublinear Reinforcement Learning

Aug 25, 2018
Sham Kakade, Mengdi Wang, Lin F. Yang

Figure 1 for Variance Reduction Methods for Sublinear Reinforcement Learning

* Fixed a bug of a previous version

Access Paper or Ask Questions

Improved Oracle Complexity of Variance Reduced Methods for Nonsmooth Convex Stochastic Composition Optimization

Jul 25, 2018
Tianyi Lin, Chenyou Fan, Mengdi Wang

Figure 1 for Improved Oracle Complexity of Variance Reduced Methods for Nonsmooth Convex Stochastic Composition Optimization

Figure 2 for Improved Oracle Complexity of Variance Reduced Methods for Nonsmooth Convex Stochastic Composition Optimization

* The error appears in the proof. More specifically, the proof of "section 3, Theorem 3.9" in the appendix has an error (c.f. Lemma B.4). Also some errors in "section 4, experiment". We have a new version arXiv:1806.00458

Access Paper or Ask Questions

Estimation of Markov Chain via Rank-Constrained Likelihood

Jul 19, 2018
Xudong Li, Mengdi Wang, Anru Zhang

Figure 1 for Estimation of Markov Chain via Rank-Constrained Likelihood

Figure 2 for Estimation of Markov Chain via Rank-Constrained Likelihood

Figure 3 for Estimation of Markov Chain via Rank-Constrained Likelihood

* Proceedings of the 35th International Conference on Machine Learning (ICML2018), Stockholm, Sweden, PMLR 80, 2018
* Accepted at ICML 2018

Access Paper or Ask Questions

Variance Reduced Value Iteration and Faster Algorithms for Solving Markov Decision Processes

Jun 28, 2018
Aaron Sidford, Mengdi Wang, Xian Wu, Yinyu Ye

Figure 1 for Variance Reduced Value Iteration and Faster Algorithms for Solving Markov Decision Processes

Figure 2 for Variance Reduced Value Iteration and Faster Algorithms for Solving Markov Decision Processes

Access Paper or Ask Questions

Improved Oracle Complexity for Stochastic Compositional Variance Reduced Gradient

Jun 01, 2018
Tianyi Lin, Chenyou Fan, Mengdi Wang, Michael I. Jordan

Figure 1 for Improved Oracle Complexity for Stochastic Compositional Variance Reduced Gradient

Figure 2 for Improved Oracle Complexity for Stochastic Compositional Variance Reduced Gradient

Figure 3 for Improved Oracle Complexity for Stochastic Compositional Variance Reduced Gradient

Figure 4 for Improved Oracle Complexity for Stochastic Compositional Variance Reduced Gradient

* arXiv admin note: text overlap with arXiv: 1802.02339; correct typos, improve proof and add experiments on real datasets

Access Paper or Ask Questions

Scalable Bilinear $π$ Learning Using State and Action Features

Apr 27, 2018
Yichen Chen, Lihong Li, Mengdi Wang

Access Paper or Ask Questions

Online Factorization and Partition of Complex Networks From Random Walks

Dec 11, 2017
Lin F. Yang, Vladimir Braverman, Tuo Zhao, Mengdi Wang

Figure 1 for Online Factorization and Partition of Complex Networks From Random Walks

Figure 2 for Online Factorization and Partition of Complex Networks From Random Walks

Access Paper or Ask Questions