Research map / Optimization / Distributed and compressed
Proximal SCOPE for Distributed Sparse Learning
Shenyi Zhao, Gong-Duo Zhang, Ming-Wei Li, Wu-Jun Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use pSCOPE instead of DBCD | distributed sparse learning · LR cov | time to 1e-3 suboptimal (s): 822 -> 0.32 | Improved |
| use pSCOPE instead of DBCD | distributed sparse learning · LR rcv1 | time to 1e-3 suboptimal (s): > 1000 -> 3.78 | Improved |
| use pSCOPE instead of DBCD | distributed sparse learning · Lasso cov | time to 1e-3 suboptimal (s): 81.9 -> 0.06 | Improved |
| use pSCOPE instead of DBCD | distributed sparse learning · Lasso rcv1 | time to 1e-3 suboptimal (s): > 1000 -> 3.09 | Improved |
| use pSCOPE instead of FISTA, DFAL, mOWL-QN, AsyProx-SVRG, PROXCOCOA+ | distributed sparse learning · cov, rcv1, avazu, kdd2012 | objective value minus minimum vs time | Improved |
| increase workers p from 1 to 8 | pSCOPE · cov, rcv1, avazu, kdd2012 LR | speedup | Improved |
| uniform partition pi1 instead of best partition pi* (whole data on each worker) (ablation) | pSCOPE · cov, rcv1 | convergence | No clear effect |
| skewed partitions pi2 or pi3 instead of uniform partition pi1 (ablation) | pSCOPE · cov, rcv1 | convergence | Hurt |
Closest papers by meaning
- An efficient distributed learning algorithm based on effective local functional approximations (JMLR 2018 · similarity 0.87)
- CoCoA: A General Framework for Communication-Efficient Distributed Optimization (JMLR 2018 · similarity 0.86)
- Distributed Machine Learning with Sparse Heterogeneous Data (NeurIPS 2021 · similarity 0.86)
- Efficient Distributed Learning with Sparsity (ICML 2017 · similarity 0.85)
- A distributed block coordinate descent method for training l1 regularized linear classifiers (JMLR 2017 · similarity 0.85)
- SCOPE: Scalable Composite Optimization for Learning on Spark (AAAI 2017 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
