portfolio site templates

Preprints


  1. Optimal scheduling of entropy regulariser for continuous-time linear-quadratic reinforcement learning (with Lukasz Szpruch and Tanut Treetanthiploet)
    Submitted (2022) [Preprint]
  2. Linear convergence of a policy gradient method for finite horizon continuous time stochastic control problems (with Christoph Reisinger and Wolfgang Stockinger)
    Submitted (2022) [Preprint]
  3. Exploration-exploitation trade-off for continuous-time episodic reinforcement learning with linear-convex models (with Lukasz Szpruch and Tanut Treetanthiploet)
    Submitted (2021) [Preprint]
  4. A fast iterative PDE-based algorithm for feedback controls of nonsmooth mean-field control problems (with Christoph Reisinger and Wolfgang Stockinger)
    Submitted (2021) [Preprint]
  5. Reinforcement learning for linear-convex models with jumps via stability analysis of feedback controls (with Xin Guo and Anran Hu)
    Submitted (2021) [Preprint]
  6. Path regularity of coupled McKean-Vlasov FBSDEs (with Christoph Reisinger and Wolfgang Stockinger) (2020) [Preprint]
  7. Optimal regularity of extended mean field controls and their piecewise constant approximation (with Christoph Reisinger and Wolfgang Stockinger)
    Submitted (2020) [Preprint]
  8. A posteriori error estimates for fully coupled McKean-Vlasov forward-backward SDEs (with Christoph Reisinger and Wolfgang Stockinger)
    Submitted (2020) [Preprint]

Refereed journal publications


  1. Logarithmic regret for episodic continuous-time linear-quadratic reinforcement learning over a finite-time horizon (with Matteo Basei, Xin Guo and Anran Hu) 
    Journal of Machine Learning Research, 23 (2022), pp. 1–34  [pdf] [Preprint]
  2. Regularity and stability of feedback relaxed controls (with Christoph Reisinger)
    SIAM Journal on Control and Optimization, 59 (2021), pp. 3118–3151 [pdf] [Preprint]
  3. A penalty scheme and policy iteration for non-local HJB variational inequalities with monotone drivers (with Christoph Reisinger)
    Computers and Mathematics with Applications, 93 (2021), pp. 199-213 [pdf] [Preprint]
  4. Rectified deep neural networks overcome the curse of dimensionality for nonsmooth value functions in zero-sum games of nonlinear stiff systems (with Christoph Reisinger)
    Analysis and Applications, 18 (2020), pp. 951-999 [Preprint]
  5. A neural network based policy iteration algorithm with global $H^2$-superlinear convergence for stochastic games on domains (with Kazufumi Ito and Christoph Reisinger)
    Foundations of Computational Mathematics, 21 (2021), pp. 331–374 [pdf]
  6. Error estimates of penalty schemes for quasi-variational inequalities arising from impulse control problems (with Christoph Reisinger)
    SIAM Journal on Control and Optimization, 58 (2020), pp. 243-276 [pdf]
  7. A penalty scheme for monotone systems with interconnected obstacles: convergence and error estimates (with Christoph Reisinger)
    SIAM Journal of Numerical Analysis, 57 (2019), pp. 1625-1648 [pdf]
  8. Approximation schemes for mixed optimal stopping and control problems with nonlinear expectations and jumps (with Roxana Dumitrescu and Christoph Reisinger)
    Applied Mathematics & Optimization, 83 (2021), pp. 1387–1429 [pdf]

Refereed conference publications


  1. Understanding Deep Architectures with Reasoning Layer (with Xinshi Chen, Christoph Reisinger and Le Song)
    Advances in Neural Information Processing Systems (NeurIPS), 2020. [Preprint]
Office: COL B.100D, Columbia House





Mail: Department of Statistics,
London School of Economics,
Houghton Street,
London, WC2A 2AE