5,401 Hits in 5.3 sec

Understanding Neural Code Intelligence Through Program Simplification [article]

Md Rafiqul Islam Rabin, Vincent J. Hellendoorn, Mohammad Amin Alipour
2021 arXiv   pre-print
A wide range of code intelligence (CI) tools, powered by deep neural networks, have been developed recently to improve programming productivity and perform program analysis.  ...  We believe that SIVAND's extracted features may help understand neural CI systems' predictions and learned behavior.  ...  This sets the stage for broader use of transparency-enhancing techniques to better understand and develop neural code intelligence models. Figure 2 : 2 Workflow of Sivand.  ... 
arXiv:2106.03353v1 fatcat:3wfadhxkzrhotmdygzsqnrpr4e

General Principles for Brain Design

Brian D. Josephson
2006 AIP Conference Proceedings  
in neural hardware.  ...  Insights can be gained in regard to such issues through the study of the role played by models and representation.  ...  Note that this integration has a top-down aspect through the way that, through the models, the behaviour of wholes can be fed back to modulate 'intelligently' the behaviour of the component parts.  ... 
doi:10.1063/1.2216620 fatcat:rckdhtedezdjdhu47lblkz5hm4

The brain-machine disanalogy revisited

Bernard P. Zeigler
2002 Biosystems (Amsterdam. Print)  
These directions open new paths for a multifaceted understanding of what biological brains do and what we can learn from them.  ...  He believed that there are fundamental lessons to be learned from the structure and behavior of biological brains that we are far from understanding or have implemented in our computers.  ...  The distinguishing feature of the one-spike neural architecture is that it relies on a temporal, rather than a firing rate, code for propagating information through neural processing layers.  ... 
doi:10.1016/s0303-2647(01)00181-2 pmid:11755495 fatcat:jugikpgdkvaydnpqunyvcrlxli

Computational Intelligence for Optimization

N Ansari, E Hou
1998 Journal of the Operational Research Society  
This is essentially a method of setting up a neural network in which the stochastic neurones are replaced by determi- nistic ones, the values of which are obtained through mean field approximation.  ...  All rights reserved. 0160-5682/98 $12.00 7 N Ansari and E Hou: Computational Intelligence for Optimization RJ Vanderbei: Linear Programming: Foundations and Extensions MW Padberg and MP Rijal: Location  ... 
doi:10.1057/palgrave.jors.2600478 fatcat:7nap4raej5cedl2sh4nvv6ds7m

The Future of Neural Networks [article]

Sachin Lakra, T.V. Prasad, G. Ramakrishna
2012 arXiv   pre-print
A new model, multi/infinite dimensional neural networks, are a recent development in the area of advanced neural networks.  ...  The paper mentions a new architecture, the pulsed neural network that is being considered as the next generation of neural networks.  ...  CHARACTERISTICS OF NEURAL NETWORKS According to [1] the theoretical understanding of how cognition arises in the brain has been advanced by the understanding of artificial neural networks that display  ... 
arXiv:1209.4855v1 fatcat:oyywnru6vfbofkwtssxjppnmdm

Automated Software Vulnerability Detection Based on Hybrid Neural Network

Xin Li, Lu Wang, Yang Xin, Yixian Yang, Qifeng Tang, Yuling Chen
2021 Applied Sciences  
This paper proposes an automatic vulnerability detection framework in source code based on a hybrid neural network.  ...  A hybrid neural network model is then applied to extract high-level features of vulnerability, which learns features both from convolutional neural networks (CNNs) and recurrent neural networks (RNNs).  ...  This form makes it easy for machine learning models to understand the dependency in a program.  ... 
doi:10.3390/app11073201 fatcat:dz6mpc6qvvhmtncj4mdczuonp4

Advancements in Sensor Technology and Decision Support Intelligent Tools to Assist Smart Livestock Farming

Luis O Tedeschi, Paul L Greenwood, Ilan Halachmi
2021 Journal of Animal Science  
The integration between concept- and data-driven modeling through parallel hybridization of mechanistic and AI models will yield a hybrid intelligent mechanistic model that, along with data collection  ...  Still, it cannot explicitly explain the underlying assumptions of the intrinsic relationships in the data core because it lacks the wisdom that confers understanding and principles.  ...  For instance, Figure 3B shows a possible integration between concept-and data-driven modeling through parallel hybridization of mechanistic and learning programming paradigms, yielding a hybrid intelligent  ... 
doi:10.1093/jas/skab038 pmid:33550395 pmcid:PMC7896629 fatcat:2emxdocbyjdd7pejbargtp2qc4

An Artificial Intelligence (AI) Defect Detection Technology Based on Software Behavior Decision Tree

Xiang-zhou CHEN, Hui-xia DING, Jie ZHANG, Yang WANG, Geng ZHANG, Ya-nan WANG
2018 DEStech Transactions on Computer Science and Engineering  
At present, artificial intelligence (AI) defect detection adoptes machine learning technology to realize code scanning and semantic analysis on software defects.  ...  The traditional machine learning technology for software defect detection is generally based on algorithms such as BP neural network model, Naïve-Bayes model, and fingerprint identification model, etc.  ...  These methods are aimed at finding and preventing code defects of known types, whose basic principle is scanning program code, extracting key grammar of program, explaining its semantics, understanding  ... 
doi:10.12783/dtcse/ccnt2018/24685 fatcat:jdu5ehvoknfyhftvlvfvlkwogi

Papers by Title

2019 2019 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-TW)  
Network to Spiking Neural Network for Hardware Implementation Convolutional Neural Network for Premature Ventricular Contraction Detection Using Wavelet Fusion on Multi-Lead ECG Convolutional Sparse Coding-based  ...  Networks On the simplification of multi-focus image fusion using dictionary-based sparse representation On the Study of Shortest-path Problem on Coal-transportation Networks using Dijkstra's Algorithm  ... 
doi:10.1109/icce-tw46550.2019.8991721 fatcat:62376ymadzge3g5xomicr5tesq

Gene Identification: Classical and Computational Intelligence Approaches

S. Bandyopadhyay, U. Maulik, D. Roy
2008 IEEE Transactions on Systems Man and Cybernetics Part C (Applications and Reviews)  
In this paper, a detailed survey on the existing classical and computational intelligence based methods for gene identification is carried out.  ...  This includes a brief description of the classical and computational intelligence methods before discussing their applications to gene finding.  ...  The best gene structure is then constructed using dynamic programming to sift through the numerous possible exon assemblies.  ... 
doi:10.1109/tsmcc.2007.906066 fatcat:g7idd6l4uzboxmm3yckleyz75e

Probabilistic machine learning and artificial intelligence

Zoubin Ghahramani
2015 Nature  
The best way to understand nonparametric models is through comparison to parametric ones.  ...  from data acquired through experience.  ...  The best way to understand nonparametric models is through comparison to parametric ones.  ... 
doi:10.1038/nature14541 pmid:26017444 fatcat:sw42v3vzcraj3mhimxr4w2g6du

Handwritten Character Recognition with Artificial Neural Networks [chapter]

Stephane Kouamo, Claude Tangha
2012 Advances in Intelligent and Soft Computing  
In this research, we will study the character recognition using a new approach based on statistical calculation and artificial neural networks.  ...  This paper describes an artificial neural network (ANN)-based system that uses a word frequency database for optical character recognition (OCR) of words in Amazigh -a North African language.  ...  In this work we presented an approach based on the probabilistic and the artificial neural networks to eliminate the problems of unclassified characters increase the recognition rate (worst classification  ... 
doi:10.1007/978-3-642-28765-7_64 dblp:conf/dcai/KouamoT12 fatcat:de2ur2cxa5dahkoyum5pirztky

Foundations of Intelligence Science

Zhongzhi Shi
2011 International Journal of Intelligence Science  
More specifically, the natural intelligence and artificial intelligence should be closely interacted in Intelligence Science study, instead of separate from each other.  ...  Artificial intelligence attempts simulation, extension and expansion of human intelligence using artificial methodology and technology.  ...  Understanding how memories are stored in the brain is an essential step toward understanding ourselves.  ... 
doi:10.4236/ijis.2011.11002 fatcat:adolk4bqwvbhjaa5sv3yomxf4a

An Inclusive Ethical Design Perspective for a Flourishing Future with Artificial Intelligent Systems

2018 European Journal of Risk Regulation  
It is held that an inclusive ethical design perspective is essential for a flourishing future with artificial intelligence.  ...  AbstractsThe article provides an inclusive outlook on artificial intelligence by introducing a three-legged design perspective that includes, but also moves beyond, ethical artificial systems design to  ...  Intelligence (XAI)' (2018), available at < explainable-artificial-intelligence > , accessed 14 November 2018.  ... 
doi:10.1017/err.2018.62 fatcat:pjrd72gmrzbd5mpfxct34vh75i

Overview of soft intelligent computing technique for supercritical fluid extraction

Sitinoor Adeib Idris, Masturah Markom
2020 International Journal of Advances in Applied Sciences  
The main advantage of intelligent systems is that the predictions can be performed easily, fast, and accurate way, which physical models unable to do.  ...  This paper shares several works that have been utilizing intelligent systems for modeling and simulating the supercritical fluid extraction process.</span>  ...  There are five types which are neural networks, fuzzy systems, evolutionary computation, ideas about probabilities and swarm intelligence.  ... 
doi:10.11591/ijaas.v9.i2.pp117-124 fatcat:ha7aav4ykfdznlpu547tmt7neu
« Previous Showing results 1 — 15 out of 5,401 results