English
Related papers

Related papers: Universally Decodable Matrices for Distributed Mat…

200 papers

Distributed implementations are crucial in speeding up large scale machine learning applications. Distributed gradient descent (GD) is widely employed to parallelize the learning task by distributing the dataset across multiple workers. A…

Information Theory · Computer Science 2021-03-02 Baturalp Buyukates , Emre Ozfatura , Sennur Ulukus , Deniz Gunduz

In distributed machine learning (DML), the training data is distributed across multiple worker nodes to perform the underlying training in parallel. One major problem affecting the performance of DML algorithms is presence of stragglers.…

Information Theory · Computer Science 2021-05-14 Amogh Johri , Arti Yardi , Tejas Bodas

In distributed machine learning, a central node outsources computationally expensive calculations to external worker nodes. The properties of optimization procedures like stochastic gradient descent (SGD) can be leveraged to mitigate the…

Distributed, Parallel, and Cluster Computing · Computer Science 2023-04-19 Maximilian Egger , Serge Kas Hanna , Rawad Bitar

Gradient coding is a technique for straggler mitigation in distributed learning. In this paper we design novel gradient codes using tools from classical coding theory, namely, cyclic MDS codes, which compare favorably with existing…

Information Theory · Computer Science 2019-07-09 Netanel Raviv , Itzhak Tamo , Rashish Tandon , Alexandros G. Dimakis

We propose a novel application of coded computing to the problem of the nearest neighbor estimation using MatDot Codes [Fahim. et.al. 2017], that are known to be optimal for matrix multiplication in terms of recovery threshold under storage…

Information Theory · Computer Science 2018-11-30 Utsav Sheth , Sanghamitra Dutta , Malhar Chaudhari , Haewon Jeong , Yaoqing Yang , Jukka Kohonen , Teemu Roos , Pulkit Grover

Within distributed learning, workers typically compute gradients on their assigned dataset chunks and send them to the parameter server (PS), which aggregates them to compute either an exact or approximate version of $\nabla L$ (gradient of…

Information Theory · Computer Science 2024-11-19 Aditya Ramamoorthy , Ruoyu Meng , Vrinda S. Girimaji

Existing gradient coding schemes introduce identical redundancy across the coordinates of gradients and hence cannot fully utilize the computation results from partial stragglers. This motivates the introduction of diverse redundancies…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-09-21 Qi Wang , Ying Cui , Chenglin Li , Junni Zou , Hongkai Xiong

In this paper, we propose a new secure distributed matrix multiplication (SDMM) scheme using the inner product partitioning. We construct a scheme with a minimal number of workers and no redundancy, and another scheme with redundancy…

Information Theory · Computer Science 2024-04-29 Okko Makkonen

A major hurdle in machine learning is scalability to massive datasets. One approach to overcoming this is to distribute the computational tasks among several workers. \textit{Gradient coding} has been recently proposed in distributed…

Information Theory · Computer Science 2020-09-16 Neophytos Charalambides , Hessam Mahdavifar , Alfred O. Hero

Matrix multiplication is a fundamental computation in many scientific disciplines. In this paper, we show that novel fast matrix multiplication algorithms can significantly outperform vendor implementations of the classical algorithm and…

Distributed, Parallel, and Cluster Computing · Computer Science 2018-01-08 Austin R. Benson , Grey Ballard

In this paper we study spread codes: a family of constant-dimension codes for random linear network coding. In other words, the codewords are full-rank matrices of size (k x n) with entries in a finite field F_q. Spread codes are a family…

Information Theory · Computer Science 2012-06-08 Elisa Gorla , Felice Manganiello , Joachim Rosenthal

The widespread adoption of distributed learning to train a global model from local data has been hindered by the challenge posed by stragglers. Recent attempts to mitigate this issue through gradient coding have proved difficult due to the…

Networking and Internet Architecture · Computer Science 2023-07-26 Tingting Yang , Xinghan Wang , Jiahong Ning , Yang Yang

Coded elastic computing enables virtual machines to be preempted for high-priority tasks while allowing new virtual machines to join ongoing computation seamlessly. This paper addresses coded elastic computing for matrix-matrix…

Information Theory · Computer Science 2025-01-30 Xi Zhong , Samuel Lu , Joerg Kliewer , Mingyue Ji

Modern learning algorithms use gradient descent updates to train inferential models that best explain data. Scaling these approaches to massive data sizes requires proper distributed gradient descent schemes where distributed worker nodes…

Information Theory · Computer Science 2017-10-30 Songze Li , Seyed Mohammadreza Mousavi Kalan , A. Salman Avestimehr , Mahdi Soltanolkotabi

Large-scale machine learning and data mining applications require computer systems to perform massive matrix-vector and matrix-matrix multiplication operations that need to be parallelized across multiple nodes. The presence of straggling…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-11-01 Ankur Mallick , Malhar Chaudhari , Utsav Sheth , Ganesh Palanikumar , Gauri Joshi

In this paper, we present a novel variation of the coded matrix multiplication problem which we refer to as fully private grouped matrix multiplication (FPGMM). In FPGMM, a master wants to compute a group of matrix products between two…

Information Theory · Computer Science 2023-05-16 Lev Tauz , Lara Dolecek

Distributed-memory matrix multiplication (MM) is a key element of algorithms in many domains (machine learning, quantum physics). Conventional algorithms for dense MM rely on regular/uniform data decomposition to ensure load balance. These…

Distributed, Parallel, and Cluster Computing · Computer Science 2015-04-21 Justus A. Calvin , Edward F. Valeev

We devise achievable encoding schemes for distributed source compression for computing inner products, symmetric matrix products, and more generally, square matrix products, which are a class of nonlinear transformations. To that end, our…

Information Theory · Computer Science 2024-05-21 Derya Malak

In distributed synchronous gradient descent (GD) the main performance bottleneck for the per-iteration completion time is the slowest \textit{straggling} workers. To speed up GD iterations in the presence of stragglers, coded distributed…

Information Theory · Computer Science 2020-11-04 Baturalp Buyukates , Emre Ozfatura , Sennur Ulukus , Deniz Gunduz

We construct optimal secure coded distributed schemes that extend the known optimal constructions over fields of characteristic 0 to all fields. A serendipitous result is that we can encode \emph{all} functions over finite fields with a…

Information Theory · Computer Science 2025-04-28 Pedro Soto
‹ Prev 1 4 5 6 7 8 10 Next ›