Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Learning of Gaussian Processes in Distributed and Communication Limited Systems
Mostafa Tavassolipour, Seyed Abolfazl Motahari, Mohammad Taghi Manzuri
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 proposed dimension reduction instead of PCA when machines hold different covariances | Dimension reduction for inner-product estimation · Gaussian data and MNIST digits 6 and 7 split by class | inner-product distortion | Improved |
| use proposed dimension reduction instead of PCA when machines hold identically distributed data | Dimension reduction for inner-product estimation · Gaussian data and MNIST split uniformly | inner-product distortion | No clear effect |
| use per-symbol quantization instead of dimension reduction | Inner-product coding schemes · Synthetic 20-dim Gaussian | bits per dimension until distortion near zero: 5 vs 12.5 bits per dimension | Improved |
| quantize GP training data at 1 bit per sample instead of higher rates (ablation) | GP trained on quantized data · Synthetic 1-D GP, N=200 | posterior fidelity to true GP | Hurt |
| raise quantization rate to R>=6 bits (ablation) | GP trained on quantized data · Synthetic 1-D GP, N=200 | posterior fidelity to true GP: R >= 6 | Improved |
| transmit heavily quantized data (low rate) versus transmitting nothing (ablation) | Broadcast GP · SARCOS, KIN40K, ABALONE | regression error (SMSE) | Hurt |
| use broadcast quantized GP instead of rBCM at about 16 bits per sample | Distributed GP · SARCOS (linear kernel) | SMSE: crosses rBCM at R = 16 | Improved |
| use broadcast quantized GP instead of BCM and rBCM at about 20 bits per sample | Distributed GP · KIN40K (SE kernel) | SMSE: R around 20 | Improved |
| quantize local inducing variables instead of raw samples | Single-center GP · KIN40K | SMSE | Improved |
Closest papers by meaning
- Distributed Gaussian Mean Estimation under Communication Constraints: Optimal Rates and Communication-Efficient Algorithms (JMLR 2024 · similarity 0.88)
- Convergence Analysis of Distributed Inference with Vector-Valued Gaussian Belief Propagation (JMLR 2018 · similarity 0.86)
- Learning Gaussian Processes by Minimizing PAC-Bayesian Generalization Bounds (NeurIPS 2018 · similarity 0.86)
- Asynchronous Distributed Gaussian Process Regression (AAAI 2025 · similarity 0.85)
- Computation-Aware Gaussian Processes: Model Selection And Linear-Time Inference (NeurIPS 2024 · similarity 0.85)
- Domain Invariant Learning for Gaussian Processes and Bayesian Exploration (AAAI 2024 · similarity 0.84)
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.
