S
Sen Na
Researcher at University of Chicago
Publications - 20
Citations - 128
Sen Na is an academic researcher from University of Chicago. The author has contributed to research in topics: Rate of convergence & Nonlinear system. The author has an hindex of 7, co-authored 20 publications receiving 84 citations.
Papers
More filters
Journal ArticleDOI
Exponential Decay in the Sensitivity Analysis of Nonlinear Dynamic Programming
Sen Na,Mihai Anitescu +1 more
TL;DR: Under uniform controllability and boundedness assumptions for the problem data, it is proved that the directional derivative of the optimal state and control at time k will have exponential decay in terms of $|k-i|$ with a decay rate $\rho$ independent of the temporal horizon length.
Journal ArticleDOI
AEGCN: An Autoencoder-Constrained Graph Convolutional Network
Mingyuan Ma,Sen Na,Hongyu Wang +2 more
TL;DR: In this paper, an autoencoder-constrained graph convolutional network is proposed to solve node classification task on graph domains, where the hidden layers are constrained by an auto-encoder.
Journal ArticleDOI
Estimating differential latent variable graphical models with applications to brain connectivity
TL;DR: It is proved that given the initialization, the estimator converges linearly with a nontrivial, minimax optimal statistical error, and it is shown that the proposed nonconvex procedure outperforms existing methods.
Posted Content
Superconvergence of Online Optimization for Model Predictive Control.
Sen Na,Mihai Anitescu +1 more
TL;DR: It is proved that the one-Newton-step-per-horizon, online, lag-$L$, model predictive control algorithm for solving discrete-time, equality-constrained, nonlinear dynamic programs exhibits a behavior that is called superconvergence; that is, the tracking error with respect to the full horizon solution is not only stable for successive horizon shifts, but also decreases with increasing shift order.
Journal ArticleDOI
The graph-based behavior-aware recommendation for interactive news
TL;DR: A graph-based behavior-aware network, which simultaneously considers six different types of behaviors as well as user’s demand on the news diversity is proposed, which achieves recommending news to different users at their different levels of concentration degrees.