<div style="text-align:justify;"> In distributed AI system, the models trained on data from potentially unreliable sources can be attacked by manipulating the training data distribution by inserting ca...<div style="text-align:justify;"> In distributed AI system, the models trained on data from potentially unreliable sources can be attacked by manipulating the training data distribution by inserting carefully crafted samples into the training set, which is known as Data Poisoning. Poisoning will to change the model behavior and reduce model performance. This paper proposes an algorithm that gives an improvement of both efficiency and security for data poisoning in a distributed AI system. The past methods of active defense often have a large number of invalid checks, which slows down the operation efficiency of the whole system. While passive defense also has problems of missing data and slow detection of error source. The proposed algorithm establishes the suspect hypothesis level to test and extend the verification of data packets and estimates the risk of terminal data. It can enhance the health degree of a distributed AI system by preventing the occurrence of poisoning attack and ensuring the efficiency and safety of the system operation. </div>展开更多
Traditional AI systems are brittle in the sense that they fail miserably whenpresented with problems even slightly outside of their limited range of expertise.A powerful, extensible strategy of Distributed Artificial ...Traditional AI systems are brittle in the sense that they fail miserably whenpresented with problems even slightly outside of their limited range of expertise.A powerful, extensible strategy of Distributed Artificial Intelligence (DAI) forovercoming such bounds is to put the system in a society of systems. So theability to coordinate group activities of individuals and to communicate betweeneach other is necessary for a language describing DAI systems. Agent-orientedlanguage NUML is such a language. It is a specific kind of object-orientedlanguage. To give formal semantics to NUML, there is the problem to for-malise object-oriented programming paradigm which is still open. The theoryof higher-order rr-calculus is a concurrent computation model with sufficientcapability, which provides us a mathematical tool to do the formalization. Thispaper tries to use higher-order T-calculus to formalise NUML.展开更多
This paper presents a model for intention maintenance. It assumes that anintention is associated with a means network, and intention maintenance is aconflict resolution upon means networks. The difference between the ...This paper presents a model for intention maintenance. It assumes that anintention is associated with a means network, and intention maintenance is aconflict resolution upon means networks. The difference between the viewpointsof AI planning and intention maintenance is explained first, then a model ofintention, the conditions under which conflicts between the means occur andthe algorithms for determining the influence of removing a means awny from ameans network are presented.展开更多
文摘<div style="text-align:justify;"> In distributed AI system, the models trained on data from potentially unreliable sources can be attacked by manipulating the training data distribution by inserting carefully crafted samples into the training set, which is known as Data Poisoning. Poisoning will to change the model behavior and reduce model performance. This paper proposes an algorithm that gives an improvement of both efficiency and security for data poisoning in a distributed AI system. The past methods of active defense often have a large number of invalid checks, which slows down the operation efficiency of the whole system. While passive defense also has problems of missing data and slow detection of error source. The proposed algorithm establishes the suspect hypothesis level to test and extend the verification of data packets and estimates the risk of terminal data. It can enhance the health degree of a distributed AI system by preventing the occurrence of poisoning attack and ensuring the efficiency and safety of the system operation. </div>
文摘Traditional AI systems are brittle in the sense that they fail miserably whenpresented with problems even slightly outside of their limited range of expertise.A powerful, extensible strategy of Distributed Artificial Intelligence (DAI) forovercoming such bounds is to put the system in a society of systems. So theability to coordinate group activities of individuals and to communicate betweeneach other is necessary for a language describing DAI systems. Agent-orientedlanguage NUML is such a language. It is a specific kind of object-orientedlanguage. To give formal semantics to NUML, there is the problem to for-malise object-oriented programming paradigm which is still open. The theoryof higher-order rr-calculus is a concurrent computation model with sufficientcapability, which provides us a mathematical tool to do the formalization. Thispaper tries to use higher-order T-calculus to formalise NUML.
文摘This paper presents a model for intention maintenance. It assumes that anintention is associated with a means network, and intention maintenance is aconflict resolution upon means networks. The difference between the viewpointsof AI planning and intention maintenance is explained first, then a model ofintention, the conditions under which conflicts between the means occur andthe algorithms for determining the influence of removing a means awny from ameans network are presented.
