Research map / Detection and segmentation / CNN features
Task-aware Distributed Source Coding under Dynamic Bandwidth
Po-han Li, Sravan Kumar Ankireddy, Ruihan Zhao, Hossein Nourkhiz Mahjoub, Ehsan Moradi Pari, ufuk topcu, Sandeep P. Chinchali, Hyeji Kim
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 |
|---|---|---|---|
| flexible task-aware bandwidth allocation instead of equal split (DAE) | NDPCA · CIFAR-10 denoising | PSNR: 1.2 dB | Improved |
| flexible task-aware bandwidth allocation instead of equal split (DAE) | NDPCA · Locate and lift | success rate: 9% | Improved |
| flexible task-aware bandwidth allocation instead of equal split (DAE) | NDPCA · Airbus detection | mAP50: 14% | Improved |
| task-aware loss instead of task-agnostic reconstruction loss (ablation) | NDPCA · all three tasks | task performance | Improved |
| single NDPCA model across bandwidths vs joint autoencoder (JAE) per bandwidth | NDPCA · Locate and lift and Airbus detection | task performance | Improved |
| use NDPCA instead of separate DAE models over several bandwidths | NDPCA · CIFAR-10 denoising | storage (MB): 5 x 8.4 -> 8.3 | Improved |
| use NDPCA instead of separate DAE models over several bandwidths | NDPCA · Airbus detection | training time (hr): 4 x 11.5 -> 13.0 | Improved |
| single-bandwidth cost of NDPCA versus DAE | NDPCA · CIFAR-10 denoising | training time (hr): 0.21 -> 0.25 | Hurt |
| weighted mix of task and reconstruction loss (ablation) | NDPCA · image reconstruction vs task | task loss and reconstruction loss | Mixed |
Closest papers by meaning
- Coded Computing for Resilient Distributed Computing: A Learning-Theoretic Framework (NeurIPS 2024 · similarity 0.84)
- Mixtures of Subspaces for Bandwidth Efficient Context Parallel Training (NeurIPS 2025 · similarity 0.82)
- Towards a Better Theoretical Understanding of Independent Subnetwork Training (ICML 2024 · similarity 0.82)
- Optimal Neural Compressors for the Rate-Distortion-Perception Tradeoff (NeurIPS 2025 · similarity 0.82)
- Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers (NeurIPS 2025 · similarity 0.82)
- Subspace Networks: Scaling Decentralized Training with Communication-Efficient Model Parallelism (NeurIPS 2025 · similarity 0.82)
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.
