Publications

Refine Results

(Filters Applied) Clear All

Enforced sparse non-negative matrix factorization

May 23, 2016
  |
Conference Paper
Author:
Brendan E. Gavin
…
Published in:
30th IEEE Int. Parallel and Distributed Processing Symp., IPDPS 2016, 23-27 May 2016.
Topic:
supercomputing
R&D area:
R&D group:

Summary

Non-negative matrix factorization (NMF) is a dimensionality reduction algorithm for data that can be represented as an undirected bipartite graph. It has become a common method for generating topic models of text data because it is known to produce good results, despite its relative simplicity of implementation and ease of computation. One challenge with applying the NMF to large datasets is that intermediate matrix products often become dense, thus stressing the memory and compute elements of the underlying system. In this article, we investigate a simple but powerful modification of the alternating least squares method of determining the NMF of a sparse matrix that enforces the generation of sparse intermediate and output matrices. This method enables the application of NMF to large datasets through improved memory and compute performance. Further, we demonstrate, empirically, that this method of enforcing sparsity in the NMF either preserves or improves both the accuracy of the resulting topic model and the convergence rate of the underlying algorithm.
READ LESS

Summary

Non-negative matrix factorization (NMF) is a dimensionality reduction algorithm for data that can be represented as an undirected bipartite graph. It has become a common method for generating topic models of text data because it is known to produce good results, despite its relative simplicity of implementation and ease of...

READ MORE
Enforced sparse non-negative matrix factorization

LLMapReduce: multi-level map-reduce for high performance data analysis

May 23, 2016
  |
Conference Paper
Author:
Chansup Byun
…
Published in:
IEEE Int. Parallel and Distributed Processing Symp., IPDPS 2016, 23-27 May 2016.
Topic:
high performance computing
R&D area:
R&D group:

Summary

The map-reduce parallel programming model has become extremely popular in the big data community. Many big data workloads can benefit from the enhanced performance offered by supercomputers. LLMapReduce provides the familiar map-reduce parallel programming model to big data users running on a supercomputer. LLMapReduce dramatically simplifies map-reduce programming by providing simple parallel programming capability in one line of code. LLMapReduce supports all programming languages and many schedulers. LLMapReduce can work with any application without the need to modify the application. Furthermore, LLMapReduce can overcome scaling limits in the map-reduce parallel programming model via options that allow the user to switch to the more efficient single-program-multiple-data (SPMD) parallel programming model. These features allow users to reduce the computational overhead by more than 10x compared to standard map-reduce for certain applications. LLMapReduce is widely used by hundreds of users at MIT. Currently LLMapReduce works with several schedulers such as SLURM, Grid Engine and LSF.
READ LESS

Summary

The map-reduce parallel programming model has become extremely popular in the big data community. Many big data workloads can benefit from the enhanced performance offered by supercomputers. LLMapReduce provides the familiar map-reduce parallel programming model to big data users running on a supercomputer. LLMapReduce dramatically simplifies map-reduce programming by providing...

READ MORE
LLMapReduce: multi-level map-reduce for high performance data analysis

Generating a multiple-prerequisite attack graph

May 17, 2016
Author:
Richard P. Lippmann
…
Published in:
PATENT-9344444
Topic:
attack graphs
R&D area:
R&D group:

Summary

In one aspect, a method to generate an attack graph includes determining if a potential node provides a first precondition equivalent to one of preconditions provided by a group of preexisting nodes on the attack graph. The group of preexisting nodes includes a first state node, a first vulnerability instance node, a first prerequisite node, and a second state node. The method also includes, if the first precondition is equivalent to one of the preconditions provided by the group of preexisting nodes, coupling a current node to a preexisting node providing the precondition equivalent to the first precondition using a first edge and if the first precondition is not equivalent to one of the preconditions provided by the group of preexisting nodes, generating the potential node as a new node on the attack graph and coupling the new node to the current node using a second edge.
READ LESS

Summary

In one aspect, a method to generate an attack graph includes determining if a potential node provides a first precondition equivalent to one of preconditions provided by a group of preexisting nodes on the attack graph. The group of preexisting nodes includes a first state node, a first vulnerability instance...

READ MORE
Generating a multiple-prerequisite attack graph

A data-stream classification system for investigating terrorist threats

May 12, 2016
  |
Conference Paper
Author:
Alexia Schulz
…
Published in:
Proc. SPIE 9851, Next-Generation Analyst IV, 98510L (May 12, 2016); doi:10.1117/12.2224104.
Topic:
cyber security
R&D area:
R&D group:

Summary

The role of cyber forensics in criminal investigations has greatly increased in recent years due to the wealth of data that is collected and available to investigators. Physical forensics has also experienced a data volume and fidelity revolution due to advances in methods for DNA and trace evidence analysis. Key to extracting insight is the ability to correlate across multi-modal data, which depends critically on identifying a touch-point connecting the separate data streams. Separate data sources may be connected because they refer to the same individual, entity or event. In this paper we present a data source classification system tailored to facilitate the investigation of potential terrorist activity. This taxonomy is structured to illuminate the defining characteristics of a particular terrorist effort and designed to guide reporting to decision makers that is complete, concise, and evidence-based. The classification system has been validated and empirically utilized in the forensic analysis of a simulated terrorist activity. Next-generation analysts can use this schema to label and correlate across existing data streams, assess which critical information may be missing from the data, and identify options for collecting additional data streams to fill information gaps.
READ LESS

Summary

The role of cyber forensics in criminal investigations has greatly increased in recent years due to the wealth of data that is collected and available to investigators. Physical forensics has also experienced a data volume and fidelity revolution due to advances in methods for DNA and trace evidence analysis. Key...

READ MORE
A data-stream classification system for investigating terrorist threats

Feedback-based social media filtering tool for improved situational awareness

May 10, 2016
  |
Conference Paper
Author:
Jason R. Thornton
…
Published in:
15th Annual IEEE Int. Symp. on Technologies for Homeland Security, HST 2016, 10-12 May 2016.
Topic:
big data
R&D area:
R&D group:

Summary

This paper describes a feature-rich model of data relevance, designed to aid first responder retrieval of useful information from social media sources during disasters or emergencies. The approach is meant to address the failure of traditional keyword-based methods to sufficiently suppress clutter during retrieval. The model iteratively incorporates relevance feedback to update feature space selection and classifier construction across a multimodal set of diverse content characterization techniques. This approach is advantageous because the aspects of the data (or even the modalities of the data) that signify relevance cannot always be anticipated ahead of time. Experiments with both microblog text documents and coupled imagery and text documents demonstrate the effectiveness of this model on sample retrieval tasks, in comparison to more narrowly focused models operating in a priori selected feature spaces. The experiments also show that even relatively low feedback levels (i.e., tens of examples) can lead to a significant performance boost during the interactive retrieval process.
READ LESS

Summary

This paper describes a feature-rich model of data relevance, designed to aid first responder retrieval of useful information from social media sources during disasters or emergencies. The approach is meant to address the failure of traditional keyword-based methods to sufficiently suppress clutter during retrieval. The model iteratively incorporates relevance feedback...

READ MORE
Feedback-based social media filtering tool for improved situational awareness

Polymer dielectrics for 3D-printed RF devices in the Ka band

May 6, 2016
  |
Journal Article
Author:
Michael T. Lis
…
Published in:
Adv. Mater. Technol., Vol. 1, No. 2, May 2016, 1600027.
Topic:
3D printing
R&D area:
R&D group:

Summary

Direct-write printing allows the fabrication of centimeter-wave radio devices. Most polymer dielectric polymer materials become lossy at frequencies above 10 GHz. Presented here is a printable dielectric material with low loss in the K a band (26.5–40 GHz). This process allows the fabrication of resonator filter devices and a radio antenna.
READ LESS

Summary

Direct-write printing allows the fabrication of centimeter-wave radio devices. Most polymer dielectric polymer materials become lossy at frequencies above 10 GHz. Presented here is a printable dielectric material with low loss in the K a band (26.5–40 GHz). This process allows the fabrication of resonator filter devices and a radio...

READ MORE
Polymer dielectrics for 3D-printed RF devices in the Ka band

A key-centric processor architecture for secure computing

May 3, 2016
  |
Conference Paper
Author:
David J. Whelihan
…
Published in:
2016 IEEE Int. Symp. on Hardware-Oriented Security and Trust, HOST 2016, 3-5 May 2016.
Topic:
cryptography
R&D area:
R&D group:

Summary

We describe a novel key-centric processor architecture in which each piece of data or code can be protected by encryption while at rest, in transit, and in use. Using embedded key management for cryptographic key handling, our processor permits mutually distrusting software written by different entities to work closely together without divulging algorithmic parameters or secret program data. Since the architecture performs encryption, decryption, and key management deeply within the processor hardware, the attack surface is minimized without significant impact on performance or ease of use. The current prototype implementation is based on the Sparc architecture and is highly applicable to small to medium-sized processing loads.
READ LESS

Summary

We describe a novel key-centric processor architecture in which each piece of data or code can be protected by encryption while at rest, in transit, and in use. Using embedded key management for cryptographic key handling, our processor permits mutually distrusting software written by different entities to work closely together...

READ MORE
A key-centric processor architecture for secure computing

Storage and Database Management for Big Data

May 3, 2016
  |
Book Chapter
Author:
Albert I. Reuther
…
Published in:
Big Data: Storage, Sharing, and Security
Topic:
big data
R&D area:
R&D group:

Summary

The ability to collect and analyze large amounts of data is a growing problem within the scientific community. The growing gap between data and user calls for innovative tools that address the challenges faced by big data volume, velocity, and verity. While there has been great progress in the world of database technologies in the past few years, there are still many fundamental considerations that must be made by scientists. For example, which of the seemingly infinite technologies are the best to use for my problem? Answers to such questions require careful understanding of the technology field in addition to the types of problems that are being solved. This chapter aims to address many of the pressing questions faced by individuals interesting in using sotrage or database technologies to solve their big data problems.
READ LESS

Summary

The ability to collect and analyze large amounts of data is a growing problem within the scientific community. The growing gap between data and user calls for innovative tools that address the challenges faced by big data volume, velocity, and verity. While there has been great progress in the world...

READ MORE
Storage and Database Management for Big Data

Wafer-scale aluminum nanoplasmonic resonators with optimized metal deposition

April 28, 2016
  |
Journal Article
Author:
Vladimir Liberman
…
Published in:
ACS Photonics, Vol. 3, No. 5, 18 May 2016, pp. 796-805.
Topic:
photonics
R&D area:
R&D group:

Summary

Spectroscopic ellipsometry is demonstrated to be an effective technique for assessing the quality of plasmonic resonances within aluminum nanostructures deposited with multiple techniques. The resonance quality of nanoplasmonic aluminum arrays is shown to be strongly dependent on the method of aluminum deposition. Three-layer metal-dielectric-metal nanopillar arrays were fabricated in a complementary metal-oxide semiconductor (CMOS) facility, with the arrays of nanopillars separated from a continuous metal underlayer by a thin dielectric spacer, to provide optimum field enhancement. Nanostructures patterned in optimized aluminum, which had been deposited with a high temperature sputtering process followed by chemical mechanical planarization, display different resonance and depolarization behavior than nanostructures deposited by the more conventional evaporation process. Full plasmonic band diagrams are mapped over a wide range of incidence angles and wavelengths using spectroscopic ellipsometry, and compared for aluminum nanostructures fabricated with two methods. The resonators fabricated from optimized aluminum exhibit a narrower bandwidth of both plasmonic resonance and depolarization parameters, indicating a higher quality resonance due to a stronger localization of the electric field. The optimized wafer-scale aluminum plasmonics fabrication should provide a pathway towards better quality devices for sensing and light detection in the ultraviolet and blue parts of the spectrum.
READ LESS

Summary

Spectroscopic ellipsometry is demonstrated to be an effective technique for assessing the quality of plasmonic resonances within aluminum nanostructures deposited with multiple techniques. The resonance quality of nanoplasmonic aluminum arrays is shown to be strongly dependent on the method of aluminum deposition. Three-layer metal-dielectric-metal nanopillar arrays were fabricated in a...

READ MORE
Wafer-scale aluminum nanoplasmonic resonators with optimized metal deposition

Cryptography for Big Data security

April 27, 2016
  |
Book Chapter
Author:
Ariel T. Hamlin
…
Published in:
Chapter 10 in Big Data: Storage, Sharing, and Security, 2016, pp. 214-87.
Topic:
cryptography
R&D area:
R&D group:

Summary

This chapter focuses on state-of-the-art provably secure cryptographic techniques for protecting big data applications. We do not focus on more established, and commonly available cryptographic solutions. The goal is to inform practitioners of new techniques to consider as they develop new big data solutions rather than to summarize the current best practice for securing data.
READ LESS

Summary

This chapter focuses on state-of-the-art provably secure cryptographic techniques for protecting big data applications. We do not focus on more established, and commonly available cryptographic solutions. The goal is to inform practitioners of new techniques to consider as they develop new big data solutions rather than to summarize the current...

READ MORE
Cryptography for Big Data security
491-500 of 2315