Research

Sorted by Topics

Reinforcement Learning and Optimal Control

• Structured Policy Iteration for Linear Quadratic Regulator.
  Y. Park et al.
  Submitted to International Conference on Machine Learning (ICML).

• Optimal Operation of a Plug-in Hybrid Vehicle with Battery Thermal and Degradation Model.
  J. Kim, Y. Park, J. Fox, S. Boyd, and W. Dally.
  To appear in Proceedings of the American Control Conference (ACC), 2020.
  

• Linear Quadratic Regulator for Resource-Efficient Cloud Services.
  Y. Park, K. Mahadik, R. Rossi, G. Wu, and H. Zhao.
  Proceedings of ACM Symposium on Cloud Computing (SoCC), 2019.

• Convergent Actor-Critic under Off-policy and Function Approximation.
  H. Maei and Y. Park
  [Slides].

Convex Optimization

• Variable Metric Proximal Gradient Method with Diagonal Barzilai-Borwein Stepsize.
  Y. Park, S. Dhar, S. Boyd, and M. Shah.
  To appear in Proceedings of International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2020.
  [Arxiv (long version)], [ICASSP], [NeurIPS Workshop]

• Linear Convergence of Cyclic SAGA.
  Y. Park and E. K. Ryu.
  Optimization Letter, 2020.
  [Arxiv]

Machine Learning

• Structured Neural Inference for Undirected Graphical Model from Heterogeneous Data.
  Y. Park et al.
  In preparation to submit Neural Information Processing Systems (NeurIPS).
  

• Network Inference via the Time-Varying Graphical Lasso.
  D. Hallac, Y. Park, S. Boyd, and J. Leskovec.
  Proceedings of ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2017
  [Code]

• Learning the Network Structure of Heterogeneous Data via Pairwise Exponential Markov Random Fields.
  Y. Park, D. Hallac, S. Boyd, and J. Leskovec.
  Proceedings of International Conference on Artificial Intelligence and Statistics (AISTATS), 2017
  [Supplementary Material], [Code]

Other Topics

• Universal Loseless Compression: Context Tree Weighting. [Slides]

• Hypercontractivity, Maximal Correlation, and Non-cooperative Simulation. [Slides, Report]

• Successive Lossy Compression for Laplacian Source. [Slides, Report]