More Like this

1. ADMM-Softmax: an ADMM approach for multinomial logistic regression

   https://doi.org/10.1553/etna_vol52s214  

   Wu Fung, Samy; Tyrväinen, Sanna; Ruthotto, Lars; Haber, Eldad (January 2020, ETNA - Electronic Transactions on Numerical Analysis)
2. DQC-ADMM: Decentralized Dynamic ADMM With Quantized and Censored Communications

   https://doi.org/10.1109/TNNLS.2021.3051638  

   Liu, Yaohua; Wu, Gang; Tian, Zhi; Ling, Qing (January 2021, IEEE Transactions on Neural Networks and Learning Systems)

   null (Ed.)
3. ADMM for multiaffine constrained optimization

   https://doi.org/10.1080/10556788.2019.1683553  

   Gao, Wenbo; Goldfarb, Donald; Curtis, Frank E. (March 2020, Optimization Methods and Software)
4. Connecting Federated ADMM to Bayes

   Swaroop, Siddarth; Khan, Mohammad Emtiyaz; Doshi-Velez, Finale (January 2025, ICLR Conference)

   We provide new connections between two distinct federated learning approaches based on (i) ADMM and (ii) Variational Bayes (VB), and propose new variants by combining their complementary strengths. Specifically, we show that the dual variables in ADMM naturally emerge through the "site" parameters used in VB with isotropic Gaussian covariances. Using this, we derive two versions of ADMM from VB that use flexible covariances and functional regularisation, respectively. Through numerical experiments, we validate the improvements obtained in performance. The work shows connection between two fields that are believed to be fundamentally different and combines them to improve federated learning. 

   more » « less
5. Connecting Federated ADMM to Bayes

   Swaroop, Siddharth; Khan, Mohammad Emtiyaz; Doshi-Velez, Finale (April 2025, International Conference on Learning Representations)