Настройки

Укажите год
-

Небесная энциклопедия

Космические корабли и станции, автоматические КА и методы их проектирования, бортовые комплексы управления, системы и средства жизнеобеспечения, особенности технологии производства ракетно-космических систем

Подробнее
-

Мониторинг СМИ

Мониторинг СМИ и социальных сетей. Сканирование интернета, новостных сайтов, специализированных контентных площадок на базе мессенджеров. Гибкие настройки фильтров и первоначальных источников.

Подробнее

Форма поиска

Поддерживает ввод нескольких поисковых фраз (по одной на строку). При поиске обеспечивает поддержку морфологии русского и английского языка
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Укажите год
Укажите год
Применить Всего найдено 3. Отображено 3.
23-06-2023 дата публикации

Attack strategy extraction method based on security alarm data

Номер: CN116318929A
Автор: JIA YAN, GU ZHAOQUAN, FANG BINXING, WEI SONGXUAN, ZHANG HUAN, DU LEI, ZHANG ZHIQIANG, LIAO QING, GAO CUIYUN, LUO WENJIAN
Принадлежит:

The invention relates to the technical field of network security, in particular to an attack strategy extraction method based on security alarm data. The method comprises the following steps: S1, acquiring single-step attack information of an attacker from an alarm text; s2, constructing an attack activity sequence set; s3, constructing a candidate attack strategy; s4, constructing an attack strategy data set; s5, pre-training is carried out; s6, model training; s7, extracting an attack strategy; and S8, carrying out manual verification. According to the method, whether a candidate attack strategy of an attacker is all effective attack steps or not is judged through a training model, and the attack purpose of the attacker can be completed through combination of the attack steps; through the model, all effective attack steps of an attacker can be associated in a candidate attack strategy enumeration mode, so that an attack strategy of the attacker is formed, and a large number of association ...

Подробнее
Номер записи: 1
11-04-2023 дата публикации

Reverse attack method and attack device for black box model and storage medium

Номер: CN115952493A
Автор: LUO WENJIAN, JIA YAN, YE ZIPENG, YANG XIANGKAI
Принадлежит:

The embodiment of the invention discloses a reverse attack method and attack device for a black box model and a storage medium, and is used for the technical field of black box models, and the method comprises the steps: inputting preset training data into a black box model and a feature extraction model, and obtaining the prediction probability of the preset training data and the data features of the preset training data; training the mapping network based on the prediction probability of the preset training data and the data features; inputting a target prediction probability obtained by inputting to-be-detected data into the black box model into the trained mapping network to obtain data features of the to-be-detected data; optimizing implicit vectors in the data features of the to-be-tested data based on a preset white box model reverse attack algorithm to obtain target implicit vectors corresponding to the to-be-tested data, and obtaining target data of reverse attack according to ...

Подробнее
Номер записи: 2
07-04-2023 дата публикации

Image processing method and related equipment

Номер: CN115937607A
Автор: LUO WENJIAN, JIA YAN, KONG LINGHAO, YE ZIPENG, ZHOU QI
Принадлежит:

The embodiment of the invention provides an image processing method and related equipment. An adversarial sample can be generated and used for attacking a multi-label classification model. The method comprises the steps of obtaining an original image sample; wherein any original image sample is an image sample which is identified by a multi-label classification model and has a plurality of different image labels; inputting the original image sample into an initial confrontation sample generation algorithm to obtain an initial confrontation sample; adding disturbance to the initial confrontation sample based on a differential evolution algorithm to generate a target fitness value of a target individual in the population and an intermediate confrontation sample corresponding to the target fitness value, and determining whether the target fitness value meets a first preset sample threshold; if yes, determining the intermediate adversarial sample as a target adversarial sample, and outputting ...

Подробнее
Номер записи: 3