NDLI: Robust Decentralized Low-Rank Matrix Decomposition

Please wait, while we are loading the content...

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 8

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 7

Issue 4(Special Issue on Crowd in Intelligent Systems, Research Note/Short Paper and Regular Papers), July 2016

Introduction to the Special Issue on Crowd in Intelligent Systems

Crowdsourcing Without a Crowd: Reliable Online Species Identification Using Bayesian Models to Minimize Crowd Size

A Crowd-Powered System for Fashion Similarity Search

Rapid Low-Cost Virtual Human Bootstrapping via the Crowd

Incentives for Effort in Crowdsourcing Using the Peer Truth Serum

PPLib: Toward the Automated Generation of Crowd Computing Programs Using Process Recombination and Auto-Experimentation

Using the Crowd to Improve Search Result Ranking and the Search Experience

Crowdsourcing Empathetic Intelligence: The Case of the Annotation of EMMA Database for Emotion and Mood Recognition

Using Crowdsourcing for Scientific Analysis of Industrial Tomographic Images

CITY FEED: A Pilot System of Citizen-Sourcing for City Issue Management

Leveraging Human Computations to Improve Schematization of Spatial Relations from Imagery

A Game-Theory Approach for Effective Crowdsource-Based Relevance Assessment

Crowdsourcing Human Annotation on Web Page Structure: Infrastructure Design and Behavior-Based Quality Control

Modality-Dependent Cross-Media Retrieval

Designing Noise-Minimal Rotorcraft Approach Trajectories

STCAPLRS: A Spatial-Temporal Context-Aware Personalized Location Recommendation System

Efficient Generalized Fused Lasso and Its Applications

Multimodular Text Normalization of Dutch User-Generated Content

Robust Decentralized Low-Rank Matrix Decomposition

Telco User Activity Level Prediction with Massive Mobile Broadband Data

Coranking the Future Influence of Multiobjects in Bibliographic Network Through Mutual Reinforcement

Multitask Low-Rank Affinity Graph for Image Segmentation and Image Annotation

Issue 3(Regular Papers, Survey Papers and Special Issue on Recommender System Benchmarks), April 2016

Issue 2(Special Issue on Causal Discovery and Inference), January 2016

Issue 1, October 2015

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 6

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 5

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 4

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 3

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 2

ACM Transactions on Intelligent Systems and Technology (TIST) : Volume 1

Robust Decentralized Low-Rank Matrix Decomposition

Content Provider	ACM Digital Library
Author	Berta, Árpád Jelasity, Márk Hegedűs, István Kocsis, Levente Benczúr, András A.
Copyright Year	2016
Abstract	Low-rank matrix approximation is an important tool in data mining with a wide range of applications, including recommender systems, clustering, and identifying topics in documents. When the matrix to be approximated originates from a large distributed system, such as a network of mobile phones or smart meters, a challenging problem arises due to the strongly conflicting yet essential requirements of efficiency, robustness, and privacy preservation. We argue that although collecting sensitive data in a centralized fashion may be efficient, it is not an option when considering privacy and efficiency at the same time. Thus, we do not allow any sensitive data to leave the nodes of the network. The local information at each node (personal attributes, documents, media ratings, etc.) defines one row in the matrix. This means that all computations have to be performed at the edge of the network. Known parallel methods that respect the locality constraint, such as synchronized parallel gradient search or distributed iterative methods, require synchronized rounds or have inherent issues with load balancing, and thus they are not robust to failure. Our distributed stochastic gradient descent algorithm overcomes these limitations. During the execution, any sensitive information remains local, whereas the global features (e.g., the factor model of movies) converge to the correct value at all nodes. We present a theoretical derivation and a thorough experimental evaluation of our algorithm. We demonstrate that the convergence speed of our method is competitive while not relying on synchronization and being robust to extreme and realistic failure scenarios. To demonstrate the feasibility of our approach, we present trace-based simulations, real smartphone user behavior analysis, and tests over real movie recommender system data.
Starting Page	1
Ending Page	24
Page Count	24
File Format	PDF
ISSN	21576904
e-ISSN	21576912
DOI	10.1145/2854157
Volume Number	7
Issue Number	4
Journal	ACM Transactions on Intelligent Systems and Technology (TIST)
Language	English
Publisher	Association for Computing Machinery (ACM)
Publisher Date	2016-05-02
Publisher Place	New York
Access Restriction	One Nation One Subscription (ONOS)
Subject Keyword	Data mining Decentralized matrix factorization Decentralized recommender systems Online learning Privacy Singular value decomposition Stochastic gradient descent
Content Type	Text
Resource Type	Article
Subject	Artificial Intelligence Theoretical Computer Science

Sl.	Authority	Responsibilities	Communication Details
1	Ministry of Education (GoI), Department of Higher Education	Sanctioning Authority	https://www.education.gov.in/ict-initiatives
2	Indian Institute of Technology Kharagpur	Host Institute of the Project: The host institute of the project is responsible for providing infrastructure support and hosting the project	https://www.iitkgp.ac.in
3	National Digital Library of India Office, Indian Institute of Technology Kharagpur	The administrative and infrastructural headquarters of the project	Dr. B. Sutradhar bsutra@ndl.gov.in
4	Project PI / Joint PI	Principal Investigator and Joint Principal Investigators of the project	Dr. B. Sutradhar bsutra@ndl.gov.in Prof. Saswat Chakrabarti will be added soon
5	Website/Portal (Helpdesk)	Queries regarding NDLI and its services	support@ndl.gov.in
6	Contents and Copyright Issues	Queries related to content curation and copyright issues	content@ndl.gov.in
7	National Digital Library of India Club (NDLI Club)	Queries related to NDLI Club formation, support, user awareness program, seminar/symposium, collaboration, social media, promotion, and outreach	clubsupport@ndl.gov.in
8	Digital Preservation Centre (DPC)	Assistance with digitizing and archiving copyright-free printed books	dpc@ndl.gov.in
9	IDR Setup or Support	Queries related to establishment and support of Institutional Digital Repository (IDR) and IDR workshops	idr@ndl.gov.in

Fully distributed robust singular value decomposition

A Fast Parallel Stochastic Gradient Method for Matrix Factorization in Shared Memory Systems

Shared-memory and shared-nothing stochastic gradient descent algorithms for matrix completion

Accelerated Singular Value Decomposition (ASVD) using momentum based Gradient Descent Optimization

Bounded matrix low rank approximation.

Nonnegative matrix factorization via rank-one downdate (2008).

14-th IEEE International Conference on Peer-to-Peer Computing Fully Distributed Robust Singular Value Decomposition

Audio classification with low-rank matrix representation features

Bounded matrix factorization for recommender system

Robust Decentralized Low-Rank Matrix Decomposition

Similar Documents

Fully distributed robust singular value decomposition

A Fast Parallel Stochastic Gradient Method for Matrix Factorization in Shared Memory Systems

Shared-memory and shared-nothing stochastic gradient descent algorithms for matrix completion

Accelerated Singular Value Decomposition (ASVD) using momentum based Gradient Descent Optimization

Bounded matrix low rank approximation.

Nonnegative matrix factorization via rank-one downdate (2008).

14-th IEEE International Conference on Peer-to-Peer Computing Fully Distributed Robust Singular Value Decomposition

Audio classification with low-rank matrix representation features

Bounded matrix factorization for recommender system

Robust Decentralized Low-Rank Matrix Decomposition