Citing SCS

If you wish to cite SCS, please use any of the following:

Original paper

Main algorithm description, derivation, and initial numerical results paper.

@article{ocpb:16,
    author       = {Brendan O'Donoghue and Eric Chu and Neal Parikh and Stephen Boyd},
    title        = {Conic Optimization via Operator Splitting and Homogeneous Self-Dual Embedding},
    journal      = {Journal of Optimization Theory and Applications},
    month        = {June},
    year         = {2016},
    volume       = {169},
    number       = {3},
    pages        = {1042-1068},
    url          = {http://stanford.edu/~boyd/papers/scs.html},
}

Latest extension

The paper that derived the extension to quadratics and describes the latest version of the algorithm is available here.

@article{odonoghue:21,
    author       = {Brendan O'Donoghue},
    title        = {Operator Splitting for a Homogeneous Embedding of the Linear Complementarity Problem},
    journal      = {{SIAM} Journal on Optimization},
    month        = {August},
    year         = {2021},
    volume       = {31},
    issue        = {3},
    pages        = {1999-2023},
}

Software

If you need to cite a particular version of the SCS software (e.g., for replication purposes) the latest version can be cited as:

@misc{scs,
    author       = {Brendan O'Donoghue and Eric Chu and Neal Parikh and Stephen Boyd},
    title        = {{SCS}: Splitting Conic Solver, version 3.2.7},
    howpublished = {\url{https://github.com/cvxgrp/scs}},
    month        = nov,
    year         = 2023
}

Anderson Acceleration

The acceleration scheme we use is described in the paper.

@article{aa2020,
  title={Globally Convergent {type--I} {A}nderson Acceleration for Non-Smooth Fixed-Point Iterations},
  author={Junzi Zhang and Brendan O'Donoghue and Stephen Boyd},
  journal={{SIAM} Journal on Optimization},
  volume={30},
  number={4},
  pages={3170--3197},
  year={2020}
}