No article text was recorded for this version.
error feedback
Version 2 of 2458 papers789 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Communication Compression for Distributed Learning without Control Variates
- Blog: ErrorCompensatedX: error compensation for variance reduced algorithms
- Blog: Training Faster with Compressed Gradient
- Blog: Step-Ahead Error Feedback for Distributed Training with Compressed Gradient
- Blog: MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence
- Blog: Rethinking gradient sparsification as total error minimization
- Blog: EF21-P and Friends: Improved Theoretical Communication Complexity for Distributed Optimization with Bidirectional Compression
- Blog: Rethinking gradient sparsification as total error minimization
- Blog: Compressed Gradient Tracking for Decentralized Optimization with Linear Convergence
- Blog: Detached Error Feedback for Distributed SGD with Random Sparsification
- Blog: A Double Residual Compression Algorithm for Efficient Distributed Learning
- Blog: DoubleSqueeze: Parallel Stochastic Gradient Descent with Double-pass Error-Compensated Compression
- Blog: Tight analyses of first-order methods with error feedback
- Blog: Communication-Efficient Distributed Learning with Local Immediate Error Compensation
- Blog: Adaptive Compression for Communication-Efficient Distributed Training
- Blog: CSER: Communication-efficient SGD with Error Reset
- Blog: Compressed Gradient Methods With Hessian-Aided Error Compensation
- Blog: Analysis of Error Feedback in Federated Non-Convex Optimization with Biased Compression: Fast Convergence and Partial Participation
- Blog: Error Feedback Fixes SignSGD and other Gradient Compression Schemes
- Blog: Compressed Gradient Methods With Hessian-Aided Error Compensation
- Blog: EF21: A New, Simpler, Theoretically Better, and Practically Faster Error Feedback
- Blog: Contractive error feedback for gradient compression
- Blog: Stochastic Controlled Averaging for Federated Learning with Communication Compression
- Blog: A Double Residual Compression Algorithm for Efficient Distributed Learning
- Blog: EControl: Fast Distributed Optimization with Compression and Error Control
- Blog: Step-Ahead Error Feedback for Distributed Training with Compressed Gradient
- Blog: Low-Rank Gradient Compression with Error Feedback for MIMO Wireless Federated Learning
- Blog: Activations and Gradients Compression for Model-Parallel Training
- Blog: Decentralization Meets Quantization
- Blog: Linearly Converging Error Compensated SGD
- Blog: Analysis of Error Feedback in Federated Non-Convex Optimization with Biased Compression
- Blog: Matching centralized learning performance via compressed decentralized learning with error feedback
- Blog: SparDL: Distributed Deep Learning Training with Efficient Sparse Communication
- Blog: Error Compensated Quantized SGD and its Applications to Large-scale Distributed Optimization
- Blog: Efficient Wireless Federated Learning via Low-Rank Gradient Factorization
- Blog: Differentially Quantized Gradient Methods
- Blog: Communication Compression for Distributed Nonconvex Optimization
- Blog: Decentralized Stochastic Optimization and Gossip Algorithms with Compressed Communication
- Blog: Tight analyses of first-order methods with error feedback
- Blog: Distributed Learning With Sparsified Gradient Differences
- Blog: Differential Error Feedback for Communication-Efficient Decentralized Learning
- Blog: MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence
- Blog: Linear Convergent Decentralized Optimization with Compression
- Blog: LoCo: Low-Bit Communication Adaptor for Large-scale Model Training
- Blog: Error Compensated Quantized SGD and its Applications to Large-scale Distributed Optimization
- Blog: Communication-efficient Vertical Federated Learning via Compressed Error Feedback
- Blog: CSER: Communication-efficient SGD with Error Reset
- Blog: Communication-Efficient Distributed Blockwise Momentum SGD with Error-Feedback
- Blog: Accelerated Sparsified SGD with Error Feedback
- Blog: Escaping Saddle Points in Heterogeneous Federated Learning via Distributed SGD with Communication Compression
- Blog: Activations and Gradients Compression for Model-Parallel Training
- Blog: Communication-Efficient Adaptive Federated Learning
- Blog: Error Compensated Loopless SVRG, Quartz, and SDCA for Distributed Optimization
- Blog: Communication-Efficient Adaptive Federated Learning
- Blog: Communication-Efficient Stochastic Gradient Descent Ascent with Momentum Algorithms
- Blog: On Communication Compression for Distributed Optimization on Heterogeneous Data
- Blog: On the Convergence of Memory-Based Distributed SGD
- Blog: CFedAvg: Achieving Efficient Communication and Fast Convergence in Non-IID Federated Learning
- Blog: Clip21: Error Feedback for Gradient Clipping
- Blog: Greedy Low-Rank Gradient Compression for Distributed Learning with Convergence Guarantees
Showing 60 of 458 papers.
History & citation
Current · Permanent link
- Sparse Gradient Compression for Distributed Training merged into this direction (56% overlap)
- Linear Speedup in Compressed Distributed Optimization merged into this direction (79% overlap)
- Error Compensated Gradient Sparsification for Distributed Learning merged into this direction (100% overlap)
- Accelerated Compressed Gradient Descent for Distributed Optimization merged into this direction (67% overlap)
- quantization error accumulation split from this direction (50% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_error_feedback_4e5457c7,
title = {error feedback},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/error-feedback/rdr_4e5457c7e450494d806c587232cca40f},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 2, frozen text):
@misc{lacuna_error_feedback_4e5457c7_v2,
title = {error feedback},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/error-feedback/rdr_4e5457c7e450494d806c587232cca40f/v/2},
note = {Version 2, frozen 2026-10-04. Accessed 2026-11-19},
}