1,760 Hits in 3.1 sec

Neural Execution Engines: Learning to Execute Subroutines [article]

Yujun Yan, Kevin Swersky, Danai Koutra, Parthasarathy Ranganathan, Milad Hashemi
2020 arXiv   pre-print
A significant effort has been made to train neural networks that replicate algorithmic reasoning, but they often fail to learn the abstract concepts underlying these algorithms.  ...  First, we observe that transformer-based sequence-to-sequence models can learn subroutines like sorting a list of numbers, but their performance rapidly degrades as the length of lists grows beyond those  ...  Execution Engine min_element append() sorted_list data learned_mask Neural Execution Engine partially_sorted_data reshape() data learned_mask end-start Neural Execution Engine  ... 
arXiv:2006.08084v3 fatcat:pqpwd3e2ubcfjan4vxpskv54i4

Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware [article]

Florian Tramèr, Dan Boneh
2019 arXiv   pre-print
This paper initiates the study of high performance execution of Deep Neural Networks (DNNs) in TEEs by efficiently partitioning DNN computations between trusted and untrusted devices.  ...  As Machine Learning (ML) gets applied to security-critical or sensitive domains, there is a growing need for integrity and privacy for outsourced ML computations.  ...  If F belongs to the server, C learns no more about F than what is revealed by y = F (x). 1 TRUSTED EXECUTION ENVIRONMENTS (TEES), INTEL SGX, AND A STRONG BASELINE Trusted Execution Environments (TEE)  ... 
arXiv:1806.03287v2 fatcat:bltsrqkdizc45clfer7durbsbq

Hunting for metamorphic engines

Wing Wong, Mark Stamp
2006 Journal in Computer Virology  
We define a similarity index and use it to precisely quantify the degree of metamorphism that each generator produces.  ...  They used three learning algorithms to train a set of classifiers on some publicly available malicious and benign executables.  ...  Machine learning techniques Various researchers have attempted to use machine learning techniques to perform heuristic analysis of metamorphic viruses.  ... 
doi:10.1007/s11416-006-0028-7 fatcat:a6oasw6fpzfyvj4s5ulraobr4y

The AGINAO Self-Programming Engine

Wojciech Skaba
2013 Journal of Artificial General Intelligence  
The dynamical and open-ended cognitive engine of the robot is represented by an embedded and multi-threaded control program, that is self-crafted rather than hand-crafted, and is executed on a simulated  ...  The data from the robot's sensory devices supplies the training samples for the machine learning methods, while the commands sent to actuators enable testing hypotheses and getting a feedback.  ...  First, the reinforcement-learning values, related to the terminated execution, are updated.  ... 
doi:10.2478/v10229-011-0018-0 fatcat:s3kgr5anyvaw3gvm6my52c7pre


2001 Handbook of Software Engineering and Knowledge Engineering  
Second, we will only report mature machine learning methods; i.e. those methods which do not require highly specialized skills to execute.  ...  Software engineers can use machine learners to simplify systems development.  ...  , see the Reynolds chapter in this volume on evolutionary programming and other work on learning knowledge from data [28] ; artificial neural nets [11] ; an excellent review on data mining [22] ; and  ... 
doi:10.1142/9789812389718_0035 fatcat:td656rynrngq7g32aelrdjezvy

Mind and autonomy in engineered biosystems

O.G. Clark, R. Kok, R. Lacroix
1999 Engineering applications of artificial intelligence  
Examples refer primarily to engineered ecosystems combined with technological control networks (ecocyborgs).  ...  A signi®cantly autonomous biosystem must be engineered to possess particular sets of computational abilities (faculties).  ...  Appreciation is expressed to M.C. Stacey and P.C. Clark for reviewing drafts of this article.  ... 
doi:10.1016/s0952-1976(99)00010-x fatcat:4nqvrfkrubcezmlbt574o4wclq

Quest For Convergence Solution Using Hybrid Genetic Algorithm Trained Neural Network Model For Metamorphic Malware Detection

Arnold Adimabua Ojugo, Chris Obaro Obruche, Andrew Okonji Eboka
2021 ARRUS Journal of Engineering and Technology  
Study proposes genetic algorithm trained neural net model to accurately classify credit card transactions.  ...  Perpetrated on customers, it ranges from employees' internal abuse to large fraud via high-value contracts cum control breaches that impose serious consequences to biz.  ...  Its engine reads in a virus executable, locates code to be transformed using its locate_own_code module.  ... 
doi:10.35877/jetech613 fatcat:frhqdx6pajeujc3ia6gt4axnka

Optimization of Classroom Illumination System Based on Neural Network Algorithm

Hongwu ZENG
2018 Light & Engineering  
fatigue and promote the learning efficiency.  ...  The optimization of classroom illumination system based on neural network algorithm was studied in order to optimize the classroom illumination system, which can effectively relieve the students С visual  ...  Besides the initialization and pin definition of the system, other subroutines of the system are executed repeatedly according to the workflow of the system, and each sub program will not interfere with  ... 
doi:10.33383/2018-138 fatcat:uu6tj4yqk5c2xc5nnejvatymqe

Pipelined HAC Estimation Engines for Multivariate Time Series

Ce Guo, Wayne Luk
2014 Journal of Signal Processing Systems  
This paper describes a pipeline-friendly HAC estimation algorithm derived from a mathematical specification, by applying transformations to eliminate conditionals, to parallelise arithmetic, and to promote  ...  One experimental system achieves up to 12 times speedup over an optimised software system on 12 CPU cores.  ...  Acknowledgments The authors would like to thank the anonymous reviewers for their constructive comments.  ... 
doi:10.1007/s11265-014-0897-9 fatcat:llewfkf3gfhrbcigbsga2t2fwe

Ultra artificial intelligence (UAI) engineering for robotics violence control, detect and corrective

Sadique Shaikh
2018 International Robotics & Automation Journal  
Acknowledgements The I really thankful to my wife Safeena Shaikh for her moral support my son Md.  ...  These alternate engineering aspects become more to most complicated and challenging as moving from Alt-1 to Alt-4 as well as must need to engineer when robotic violence in Humanoid become more to most  ...  and malfunctions possible when Humanoid become most advanced Robot with self-learning and programming.  ... 
doi:10.15406/iratj.2018.04.00129 fatcat:nkulpkvf4fggthw2dbxsy4ahmq

Bots in software engineering: a systematic mapping study

Sivasurya Santhanam, Tobias Hecking, Andreas Schreiber, Stefan Wagner
2022 PeerJ Computer Science  
Bots have emerged from research prototypes to deployable systems due to the recent developments in machine learning, natural language processing and understanding techniques.  ...  Hence it is significant to categorize bots in software engineering through analyzing why, what and how the bots are applied in software engineering.  ...  Neural QA model trained with encoder-decoder Recurrent neural network (RNN) has been adopted for software engineering chatbots.  ... 
doi:10.7717/peerj-cs.866 pmid:35494879 pmcid:PMC9044364 fatcat:xmkvbhry3bg77g3c4l333aavfa

Requirements for guidelines systems: implementation challenges and lessons from existing software-engineering efforts

Hemant Shah, Raymond D Allard, Robert Enberg, Ganesh Krishnan, Patricia Williams, Prakash M Nadkarni
2012 BMC Medical Informatics and Decision Making  
During such an analysis, study of examples of existing, software-engineering efforts in non-biomedical fields can provide useful signposts to the implementer of a clinical guideline system.  ...  support systems in general, we have incorporated additional requirements related to production-system robustness and functionality from publications in the business workflow domain, in addition to drawing  ...  Workflows are generally executed in interpreted mode by a runtime workflow-execution engine.  ... 
doi:10.1186/1472-6947-12-16 pmid:22405400 pmcid:PMC3342141 fatcat:ynpbzfruwffivbbalu3du37yhq

Outbreak Prediction of Dengue and Hepatitis A

Miron T Manuel
2020 International Journal for Research in Applied Science and Engineering Technology  
We developed a model 'OUTBREAK PREDICTION OF DENGUE AND HEPATITIS A USING MACHINE LEARNING' which can be put to use as an early warning tool.  ...  In this study we use two data mining classification algorithms Support Vector Machine (SVM) and Artificial Neural Network (ANN) are used to forecast or predict an outbreak.  ...  Artificial Neural Network is proficient of learning, which is done by changing weight values. If the network gives rise to a good output, there is no need to modify the weights value.  ... 
doi:10.22214/ijraset.2020.30562 fatcat:injx6fpn6nfnbd5m3z3doubu5e

Instruction set architectural guidelines for embedded packet-processing engines

Mostafa E. Salehi, Sied Mehdi Fakhraie, Amir Yazdanbakhsh
2012 Journal of systems architecture  
Similar to other embedded processors such as media processors, packet-processing engines are deployed in embedded applications, where cost and power are as important as performance.  ...  This paper presents instruction set architectural guidelines for improving general-purpose embedded processors to optimally accommodate packet-processing applications.  ...  There are four common types for control flow instructions: conditional branches, jumps, subroutine calls, and returns from subroutines.  ... 
doi:10.1016/j.sysarc.2012.02.004 fatcat:jzaqkm5jejdpxmlory3ee4ljgy

The R Language: An Engine for Bioinformatics and Data Science

Federico M. Giorgi, Carmine Ceraolo, Daniele Mercatelli
2022 Life  
It currently ranks among the top 10 most popular languages worldwide, and its community has produced tens of thousands of extensions and packages, with scopes ranging from machine learning to transcriptome  ...  Overall, we hope to provide both a complete snapshot of R today and a practical compendium of the major features and applications of this programming language.  ...  Acknowledgments: We would like to thank Francesco Licausi and Beatrice Giuntoli for their scientific support, and Mariangela Santorsola for the fruitful discussions on code editors.  ... 
doi:10.3390/life12050648 fatcat:vyktde5d5vf4laughw7iwxlth4
« Previous Showing results 1 — 15 out of 1,760 results