Research map / Architectures and efficiency / RNNs and normalization
Berrut Approximated Coded Computing: Straggler Resistance Beyond Polynomial Computing
Tayyebeh Jahani-Nezhad, Mohammad Ali Maddah-Ali
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 Chebyshev points instead of equidistant points in BACC (ablation) | BACC · Polynomial functions, N=500, K=20, deg f=25 (simulation) | relative error: reduces the expectation of the relative error by an order of magnitude | Improved |
| add more non-straggling workers' results | BACC · Polynomial functions, simulation | relative error | Improved |
| use BACC with fewer than (K-1)deg f+1 workers | BACC vs Lagrange coded computing · Polynomial functions, simulation | feasibility | Improved |
| apply BACC to non-polynomial f=x sin x | BACC · x sin x, N=60, s=20, K=20 (simulation) | relative error | Improved |
| use BACC coded data instead of data replication in distributed LeNet training | BACC · MNIST, Fashion-MNIST, CIFAR-10 (N=3,s=1; N=5,s=2) | test accuracy vs wall-clock time | Mixed |
| use BACC instead of no data redundancy in distributed training | BACC · MNIST, Fashion-MNIST, CIFAR-10 | final test accuracy | Improved |
Closest papers by meaning
- Coded Computing for Resilient Distributed Computing: A Learning-Theoretic Framework (NeurIPS 2024 · similarity 0.87)
- Sequential Gradient Coding For Straggler Mitigation (ICLR 2023 · similarity 0.87)
- Communication-Computation Efficient Gradient Coding (ICML 2018 · similarity 0.85)
- Leveraging partial stragglers within gradient coding (NeurIPS 2024 · similarity 0.85)
- Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers (NeurIPS 2025 · similarity 0.85)
- Towards a Better Theoretical Understanding of Independent Subnetwork Training (ICML 2024 · similarity 0.83)
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.
