A Markov Chain Theory Approach to Characterizing the Minimax Optimality of Stochastic Gradient Descent (for Least Squares)

Prateek Jain 0002, Sham M. Kakade, Rahul Kidambi, Praneeth Netrapalli, Venkata Krishna Pillutla, Aaron Sidford. A Markov Chain Theory Approach to Characterizing the Minimax Optimality of Stochastic Gradient Descent (for Least Squares). In Satya V. Lokam, R. Ramanujam, editors, 37th IARCS Annual Conference on Foundations of Software Technology and Theoretical Computer Science, FSTTCS 2017, December 11-15, 2017, Kanpur, India. Volume 93 of LIPIcs, Schloss Dagstuhl - Leibniz-Zentrum fuer Informatik, 2017. [doi]

@inproceedings{JainKKNPS17,
  title = {A Markov Chain Theory Approach to Characterizing the Minimax Optimality of Stochastic Gradient Descent (for Least Squares)},
  author = {Prateek Jain 0002 and Sham M. Kakade and Rahul Kidambi and Praneeth Netrapalli and Venkata Krishna Pillutla and Aaron Sidford},
  year = {2017},
  doi = {10.4230/LIPIcs.FSTTCS.2017.2},
  url = {http://dx.doi.org/10.4230/LIPIcs.FSTTCS.2017.2},
  researchr = {https://researchr.org/publication/JainKKNPS17},
  cites = {0},
  citedby = {0},
  booktitle = {37th IARCS Annual Conference on Foundations of Software Technology and Theoretical Computer Science, FSTTCS 2017, December 11-15, 2017, Kanpur, India},
  editor = {Satya V. Lokam and R. Ramanujam},
  volume = {93},
  series = {LIPIcs},
  publisher = {Schloss Dagstuhl - Leibniz-Zentrum fuer Informatik},
  isbn = {978-3-95977-055-2},
}