Table | Card | RUSMARC | |
Allowed Actions: –
Action 'Read' will be available if you login or access site from another network
Action 'Download' will be available if you login or access site from another network
Group: Anonymous Network: Internet |
Annotation
В данной работе изложена сущность подхода к применению нейронных сетей глубокого обучения. Даны общие понятия rtf-объектов, LSTM сетей и модели Se-quence to sequence. Проведён сбор обучающего датасета, а также обучение нейрон-ной сети на этом наборе. Разработан алгоритм генерации rtf-объектов и проведён анализ его эффективности. Осуществлён фаззинг rtf-файлов с применением разрабо-танного подхода.
In the given work the essence of the approach to the use of deep learning neural networks. Given the general concepts of rtf-objects, LSTM networks and models of Se-quence to sequence. A training dataset was collected, and a neural network was trained on this set. An algorithm for generating rtf objects has been developed and its effectiveness has been analyzed. Fuzzing rtf files using the developed approach was performed.
Document access rights
Network | User group | Action | ||||
---|---|---|---|---|---|---|
ILC SPbPU Local Network | All | |||||
Internet | Authorized users SPbPU | |||||
Internet | Anonymous |
Usage statistics
Access count: 76
Last 30 days: 0 Detailed usage statistics |