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

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Qianxiao Li

Qianxiao Li

QROSS: QUBO Relaxation Parameter Optimisation via Learning Solver Surrogates


Mar 19, 2021
Tian Huang, Siong Thye Goh, Sabrish Gopalakrishnan, Tao Luo, Qianxiao Li, Hoong Chuin Lau

Figure 1 for QROSS: QUBO Relaxation Parameter Optimisation via Learning Solver Surrogates
Figure 2 for QROSS: QUBO Relaxation Parameter Optimisation via Learning Solver Surrogates
Figure 3 for QROSS: QUBO Relaxation Parameter Optimisation via Learning Solver Surrogates
Figure 4 for QROSS: QUBO Relaxation Parameter Optimisation via Learning Solver Surrogates
* 17 pages 
  Access Paper or Ask Questions

Towards Robust Neural Networks via Close-loop Control


Feb 03, 2021
Zhuotong Chen, Qianxiao Li, Zheng Zhang

Figure 1 for Towards Robust Neural Networks via Close-loop Control
Figure 2 for Towards Robust Neural Networks via Close-loop Control
Figure 3 for Towards Robust Neural Networks via Close-loop Control
Figure 4 for Towards Robust Neural Networks via Close-loop Control
* Published as a conference paper at ICLR 2021 
  Access Paper or Ask Questions

Amata: An Annealing Mechanism for Adversarial Training Acceleration


Dec 15, 2020
Nanyang Ye, Qianxiao Li, Xiao-Yun Zhou, Zhanxing Zhu

Figure 1 for Amata: An Annealing Mechanism for Adversarial Training Acceleration
Figure 2 for Amata: An Annealing Mechanism for Adversarial Training Acceleration
Figure 3 for Amata: An Annealing Mechanism for Adversarial Training Acceleration
Figure 4 for Amata: An Annealing Mechanism for Adversarial Training Acceleration
* accepted by AAAI 
  Access Paper or Ask Questions

A Data Driven Method for Computing Quasipotentials


Dec 13, 2020
Bo Lin, Qianxiao Li, Weiqing Ren

Figure 1 for A Data Driven Method for Computing Quasipotentials
Figure 2 for A Data Driven Method for Computing Quasipotentials
Figure 3 for A Data Driven Method for Computing Quasipotentials
Figure 4 for A Data Driven Method for Computing Quasipotentials

  Access Paper or Ask Questions

Optimising Stochastic Routing for Taxi Fleets with Model Enhanced Reinforcement Learning


Oct 22, 2020
Shen Ren, Qianxiao Li, Liye Zhang, Zheng Qin, Bo Yang

Figure 1 for Optimising Stochastic Routing for Taxi Fleets with Model Enhanced Reinforcement Learning
Figure 2 for Optimising Stochastic Routing for Taxi Fleets with Model Enhanced Reinforcement Learning
Figure 3 for Optimising Stochastic Routing for Taxi Fleets with Model Enhanced Reinforcement Learning
Figure 4 for Optimising Stochastic Routing for Taxi Fleets with Model Enhanced Reinforcement Learning
* 24 pages, comments welcome 
  Access Paper or Ask Questions

OnsagerNet: Learning Stable and Interpretable Dynamics using a Generalized Onsager Principle


Oct 04, 2020
Haijun Yu, Xinyuan Tian, Weinan E, Qianxiao Li

Figure 1 for OnsagerNet: Learning Stable and Interpretable Dynamics using a Generalized Onsager Principle
Figure 2 for OnsagerNet: Learning Stable and Interpretable Dynamics using a Generalized Onsager Principle
Figure 3 for OnsagerNet: Learning Stable and Interpretable Dynamics using a Generalized Onsager Principle
Figure 4 for OnsagerNet: Learning Stable and Interpretable Dynamics using a Generalized Onsager Principle
* 37 pages, 17 figures 
  Access Paper or Ask Questions

On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis


Sep 16, 2020
Zhong Li, Jiequn Han, Weinan E, Qianxiao Li

Figure 1 for On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis
Figure 2 for On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis
Figure 3 for On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis
Figure 4 for On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis

  Access Paper or Ask Questions

Inverse design of crystals using generalized invertible crystallographic representation


May 15, 2020
Zekun Ren, Juhwan Noh, Siyu Tian, Felipe Oviedo, Guangzong Xing, Qiaohao Liang, Armin Aberle, Yi Liu, Qianxiao Li, Senthilnath Jayavelu, Kedar Hippalgaonkar, Yousung Jung, Tonio Buonassisi

Figure 1 for Inverse design of crystals using generalized invertible crystallographic representation
Figure 2 for Inverse design of crystals using generalized invertible crystallographic representation
Figure 3 for Inverse design of crystals using generalized invertible crystallographic representation
Figure 4 for Inverse design of crystals using generalized invertible crystallographic representation

  Access Paper or Ask Questions

Optimization in Machine Learning: A Distribution Space Approach


Apr 18, 2020
Yongqiang Cai, Qianxiao Li, Zuowei Shen

Figure 1 for Optimization in Machine Learning: A Distribution Space Approach
Figure 2 for Optimization in Machine Learning: A Distribution Space Approach
Figure 3 for Optimization in Machine Learning: A Distribution Space Approach
Figure 4 for Optimization in Machine Learning: A Distribution Space Approach
* 26 pages, 12 figures 
  Access Paper or Ask Questions

Collaborative Inference for Efficient Remote Monitoring


Feb 12, 2020
Chi Zhang, Yong Sheng Soh, Ling Feng, Tianyi Zhou, Qianxiao Li

Figure 1 for Collaborative Inference for Efficient Remote Monitoring
Figure 2 for Collaborative Inference for Efficient Remote Monitoring
Figure 3 for Collaborative Inference for Efficient Remote Monitoring
Figure 4 for Collaborative Inference for Efficient Remote Monitoring

  Access Paper or Ask Questions

Deep Learning via Dynamical Systems: An Approximation Perspective


Dec 22, 2019
Qianxiao Li, Ting Lin, Zuowei Shen


  Access Paper or Ask Questions

Computing Committor Functions for the Study of Rare Events Using Deep Learning


Jun 14, 2019
Qianxiao Li, Bo Lin, Weiqing Ren

Figure 1 for Computing Committor Functions for the Study of Rare Events Using Deep Learning
Figure 2 for Computing Committor Functions for the Study of Rare Events Using Deep Learning
Figure 3 for Computing Committor Functions for the Study of Rare Events Using Deep Learning
Figure 4 for Computing Committor Functions for the Study of Rare Events Using Deep Learning

  Access Paper or Ask Questions

Distributed Optimization for Over-Parameterized Learning


Jun 14, 2019
Chi Zhang, Qianxiao Li

Figure 1 for Distributed Optimization for Over-Parameterized Learning
Figure 2 for Distributed Optimization for Over-Parameterized Learning
Figure 3 for Distributed Optimization for Over-Parameterized Learning
Figure 4 for Distributed Optimization for Over-Parameterized Learning

  Access Paper or Ask Questions

Stochastic Modified Equations and Dynamics of Stochastic Gradient Algorithms I: Mathematical Foundations


Nov 05, 2018
Qianxiao Li, Cheng Tai, Weinan E

Figure 1 for Stochastic Modified Equations and Dynamics of Stochastic Gradient Algorithms I: Mathematical Foundations
Figure 2 for Stochastic Modified Equations and Dynamics of Stochastic Gradient Algorithms I: Mathematical Foundations
Figure 3 for Stochastic Modified Equations and Dynamics of Stochastic Gradient Algorithms I: Mathematical Foundations
Figure 4 for Stochastic Modified Equations and Dynamics of Stochastic Gradient Algorithms I: Mathematical Foundations

  Access Paper or Ask Questions

On the Convergence and Robustness of Batch Normalization


Sep 29, 2018
Yongqiang Cai, Qianxiao Li, Zuowei Shen

Figure 1 for On the Convergence and Robustness of Batch Normalization
Figure 2 for On the Convergence and Robustness of Batch Normalization
Figure 3 for On the Convergence and Robustness of Batch Normalization
Figure 4 for On the Convergence and Robustness of Batch Normalization

  Access Paper or Ask Questions

A Mean-Field Optimal Control Formulation of Deep Learning


Jul 03, 2018
Weinan E, Jiequn Han, Qianxiao Li

* 44 pages 
  Access Paper or Ask Questions

Maximum Principle Based Algorithms for Deep Learning


Jun 02, 2018
Qianxiao Li, Long Chen, Cheng Tai, Weinan E

Figure 1 for Maximum Principle Based Algorithms for Deep Learning
Figure 2 for Maximum Principle Based Algorithms for Deep Learning
Figure 3 for Maximum Principle Based Algorithms for Deep Learning
Figure 4 for Maximum Principle Based Algorithms for Deep Learning
* Journal of Machine Learning Research 18 (2018) 1-29 
 * Published version 
  Access Paper or Ask Questions

An Optimal Control Approach to Deep Learning and Applications to Discrete-Weight Neural Networks


Jun 02, 2018
Qianxiao Li, Shuji Hao

Figure 1 for An Optimal Control Approach to Deep Learning and Applications to Discrete-Weight Neural Networks
Figure 2 for An Optimal Control Approach to Deep Learning and Applications to Discrete-Weight Neural Networks

  Access Paper or Ask Questions

Stochastic modified equations and adaptive stochastic gradient algorithms


Jun 20, 2017
Qianxiao Li, Cheng Tai, Weinan E

Figure 1 for Stochastic modified equations and adaptive stochastic gradient algorithms
Figure 2 for Stochastic modified equations and adaptive stochastic gradient algorithms
Figure 3 for Stochastic modified equations and adaptive stochastic gradient algorithms
Figure 4 for Stochastic modified equations and adaptive stochastic gradient algorithms
* Major changes including a proof of the weak approximation, asymptotic expansions and application-oriented adaptive algorithms 
  Access Paper or Ask Questions