paper-with-me

Papers

Estimation, Optimization, and Parallelism when Data is Sparse

2013-12-01 · NeurIPS 2013 12 · John Duchi, Michael. I. Jordan, Brendan Mcmahan

We study stochastic optimization problems when the \emph{data} is sparse, which is in a sense dual to the current understanding of high-dimensional statistical learning and optimization. We highlight both the difficulties---in terms of increased sample complexity that sparse data necessitates---and the potential benefits, in terms of allowing parallelism and asynchrony in the design of algorithms. Concretely, we derive matching upper and lower bounds on the minimax rate for optimization and learning with sparse data, and we exhibit algorithms achieving these rates. Our algorithms are adaptive: they achieve the best possible rate for the data observed. We also show how leveraging sparsity leads to (still minimax optimal) parallel and asynchronous algorithms, providing experimental evidence complementing our theoretical results on medium to large-scale learning tasks.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Stochastic Optimization

Similar Papers 제목 키워드 기반

Sequence Parallelism: Long Sequence Training from System Perspective

2021-05-26 · Shenggui Li, Fuzhao Xue, Chaitanya Baranwal, Yongbin Li 외

Transformer achieves promising results on various tasks. However, self-attention suffers from quadratic memory requirements with respect to the sequence length. Existing work focuses on reducing time and space complexity…

GPU

AsyncMesh: Fully Asynchronous Optimization for Data and Pipeline Parallelism

2026-01-30 · Thalaiyasingam Ajanthan, Sameera Ramasinghe, Gil Avraham, Hadi Mohaghegh Dolatabadi 외 arxiv

Data and pipeline parallelism are key strategies for scaling neural network training across distributed devices, but their high communication cost necessitates co-located computing clusters with fast interconnects, limit…

db-SP: Accelerating Sparse Attention for Visual Generative Models with Dual-Balanced Sequence Parallelism

2025-11-28 · Siqi Chen, Ke Hong, Tianchen Zhao, Ruiqi Xie 외 arxiv

Scaling Diffusion Transformer (DiT) inference via sequence parallelism is critical for reducing latency in visual generation, but is severely hampered by workload imbalance when applied to models employing block-wise spa…

A Hybrid Tensor-Expert-Data Parallelism Approach to Optimize Mixture-of-Experts Training

2023-03-11 · Siddharth Singh, Olatunji Ruwase, Ammar Ahmad Awan, Samyam Rajbhandari 외

Mixture-of-Experts (MoE) is a neural network architecture that adds sparsely activated expert blocks to a base model, increasing the number of parameters without impacting computational costs. However, current distribute…

Mixture-of-Experts

Accelerating Sparse Ternary GEMM for Quantized ML on Apple Silicon

2025-10-08 · Baraq Lipshitz, Alessio Melone, Charalampos Maraziaris, Muhammed Bilal arxiv

Sparse Ternary General Matrix-Matrix Multiplication (GEMM) remains under-optimized in existing libraries for Apple Silicon CPUs. We present a Sparse Ternary GEMM kernel optimized specifically for Apple's M-series process…