Author: SRI International
-
Neural network joint modeling via context-dependent projection
We propose a context-dependent linear projection approach which considers neighboring words. Experimental results showed that the proposed approach further improves NNJM by 0.5 BLEU for English-Iraqi Arabic translation in N-best rescoring.
-
Enabling Smart Camera Networks with Smartphone Processors
Distributed smart cameras exploit smartphone processor performance in their node communication and video metadata exchange, allowing the network to collectively reason in interpreting the scene, generating alerts, and making decisions.
-
Design-Based Research for Deeper Learning in an Online Introductory CS Course for Middle School Students
This paper describes our efforts to use the design-based research methodology to design a 6-week middle school course on the OpenEdX platform for blended in-class learning.
-
Effects of feature type, learning algorithm and speaking style for depression detection from speech
We systematically study the effects of feature type, machine learning approach, and speaking style (read versus spontaneous) on depression prediction in the AVEC-2014 evaluation corpus.
-
Reasoning About Interruption of Biological Processes
Our work is motivated by the vision of automated asking and answering of questions related to a biology textbook—a capability which requires application of abstract reasoning patterns.
-
“Systems of Assessments” for Deeper Learning of Computational Thinking in K-12
This paper argues for the need for multiple measures or “systems of assessments” that are complementary, attend to cognitive and non-cognitive aspects of CT, and contribute to a comprehensive picture of student learning.
-
AKI in Hospitalized Children: Comparing the pRIFLE, AKIN, and KDIGO Definitions
This study applied the Pediatric RIFLE, AKI Network, and Kidney Disease Improving Global Outcomes criteria to a cohort of hospitalizations to compare AKI incidence and outcomes in ICU and non-ICU pediatric populations.
-
Semantically Secure Order-Revealing Encryption: Multi-Input Functional Encryption without Obfuscation
We construct the first implementable encryption system supporting greater-than comparisons on encrypted data that provides the “best-possible” semantic security.
-
SRI Leads Research Team to Next Phase of Wendy Schmidt Ocean Health XPRIZE Competition
The health of the world’s oceans impacts all of our lives, because oceans are much like the lungs of our planet, albeit acting in reverse—consuming carbon dioxide (CO2) and releasing oxygen. The Wendy Schmidt Ocean Health XPRIZE aims to improve understanding of how CO2 emissions are acidifying the oceans through a $2 million competition to create pH sensor…
-
Soft Robots Are Reshaping the Future of Robotics
Unlike traditional robots, nature offers hard and soft structures, actuators, sensors, and grippers that achieve many of these capabilities. Soft structures and systems are also inherently safer when interacting with people.
-
ICN Hop by Hop Fragmentation
This document describes a hop-by-hop fragmentation scheme for ICN and mappings to the CCNx 1.0 and to the NDNTLV packet format.
-
Future Ready Schools Learning Guide Offers Lessons for Connected Schools
President Obama, U.S. Secretary of Education Arne Duncan, and Office of Educational Technology (OET) Director Richard Culatta recently convened “ConnectED to the Future,” a gathering of more than 100 school district superintendents and other education leaders at the White House. On the agenda: increasing broadband connectivity access to and throughout schools as well as planning for the…