Search Keywords

Results Restricted To:

https://www.temple.edu

Total Results: 8,610

RepObE: Representation Learning-Enhanced Obfuscation Encryption Modular ...

https://cis.temple.edu/~jiewu/research/publications/Publication_files/Paper%207432%20Camera%20Ready%20Version.pdf

Abstract Model inversion and adversarial attacks in se-mantic communication pose risks, such as con-tent leaks, alterations, and prediction inaccuracies, which threaten security and reliability. This pa-per introduces, from an attacker’s viewpoint, a novel framework called RepObE (Representation Learning-Enhanced Obfuscation Encryption Mod-ular Semantic Task Framework) to secure semantic ...

Embedding problems in Symplectic Topology - cst.temple.edu

https://cst.temple.edu/sites/cst/files/Templetalk19.pdf

symplectic structure is a rather elusive geometric structure that can be put on an even dimensional space. This talk will describe some basic properties of this structure, explain some fundamental results and brie y discuss some open problems.

Social Media and Firm Equity Value

https://www.fox.temple.edu/sites/fox/files/social-media-isr.pdf

Companies have increasingly advocated social media technologies to transform businesses and improve orga- nizational performance. This study scrutinizes the predictive relationships between social media and firm equity value, the relative effects of social media metrics compared with conventional online behavioral metrics, and the dynamics of these relationships. The results derived from ...

Library Search - Temple University

https://librarysearch.temple.edu/databases

Temple Libraries' official online search tool for books, media, journals, databases, government documents and more.

Directional and Explainable Serendipity Recommendation

https://cis.temple.edu/~jiewu/research/publications/Publication_files/jiang_www_2020.pdf

ABSTRACT Serendipity recommendation has attracted more and more atten-tion in recent years; it is committed to providing recommendations which could not only cater to users’ demands but also broaden their horizons. However, existing approaches usually measure user-item relevance with a scalar instead of a vector, ignoring user preference direction, which increases the risk of unrelated ...

Mini-Workshop on Software Defined Radio Using the RTL-SDR

https://sites.temple.edu/silage/files/2024/04/Mini-Workshop-On-SDR-Setup-of-SDR.pdf

SDR# is the most commonly used SDR program on Windows. We recommend it as one of the easiest to setup and use with the RTL-SDR.

doi:10.1016/j.jfineco.2004.09.004 - Sites

https://sites.temple.edu/lnaveen/files/2014/05/JFE_risktaking_2006.pdf

1. Introduction and motivation This paper provides empirical evidence of a strong causal relation between an important organizational feature, the structure of managerial compensation and corresponding incentives, and value-critical managerial decisions, specifically, those derived from both investment policy and debt policy. The primary characteristic of compensation that we consider is the ...

DeepIDPS: An Adaptive DRL-based Intrusion Detection and Prevention ...

https://cis.temple.edu/~jiewu/research/publications/Publication_files/ICC2024.pdf

Abstract—Most intrusion detection systems (IDS) are vul-nerable to novel attacks and struggle to maintain a balance between high accuracy and a low false positive rate. Furthermore, the relevant features of Distributed Denial of Service (DDoS) attacks in conventional networks may not necessarily apply to the Software-defined network (SDN) environment. Additionally, weak feature selection ...

Hamiltonian for a Charged Particle in an Electromagnetic Field

https://faculty.cst.temple.edu/~tuf43817/PHYS5701/S18/HamiltonianEM.pdf

We can derive the Hamiltonian from the Lagrangian, but it is not obvious how to build the Lagrangian when there is no obvious “potential energy” function for a charged particle in a magnetic field.

ArrayPipe: Introducing Job-Array Pipeline Parallelism for High ...

https://cis.temple.edu/~jiewu/research/publications/Publication_files/ArrayPipe-JLU-Infocom-20250520.pdf

i. A novel parallel scheme (JAP) is introduced to enable a batch of sibling jobs to form a concurrent job-array and to execute concurrently, targeting high throughput model exploration. ii. We design ArrayPipe, a framework to support JAP with low-cost job context switching within a job-array and a GPU-Host memory manager for higher training concurrency. iii. We propose a novel scheduling ...