Knowledge-guided semantic computing network
نویسندگان
چکیده
The excellent performance of deep neural networks mainly relies on the attributes dataset, such as size, diversity, completeness. However, it is usually difficult to obtain a high-qualified training dataset for some scenarios. Inspired by human visual cognition process with few sample learning, and strong robustness, we believe experience knowledge more powerful than large-scale data. To combine power data, propose knowledge-guided semantic computing network (SCN) in this paper, which constructed primary tree module an auxiliary data-driven lightweight module. can calculate classification results forward rapidly. aid higher ability. We also hinge cross-entropy loss function train SCN, enables SCN focus those misclassified samples further improve accuracy. experimental MNIST GTSRB data sets prove that achieves accuracy comparable state-of-the-art methods original samples. What more, at BIM eps = 0.3 FGSM 0.03 adversarial test samples, proposed SCN(1/4) SCN(1/8) over 75% 14% improvement CapsNet.
منابع مشابه
The Effects of Guided and Unguided Semantic Mapping on Iranian EFL Learners’ Vocabulary and Grammar Knowledge
متن کامل
ASKNet: Automated Semantic Knowledge Network
The ASKNet system is an attempt to automatically generate large scale semantic knowledge networks from natural language text. State-of-the-art language processing tools, including parsers and semantic analysers, are used to turn input sentences into fragments of semantic network. These network fragments are combined using spreading activation-based algorithms which utilise both lexical and sema...
متن کاملDNA Computing Approach to Semantic Knowledge Representation
DNA computing has a lot of potential, in terms of ability to implement a relational database with circular molecules. In this work, a new DNA-based semantic model is proposed and described theoretically for implementing DNA based memories. This model, referred to as ‘semantic model based on molecular computing’ (SMC), has the structure of a graph formed by the set of all attribute-value pairs c...
متن کاملSemantic Computing and Language Knowledge Bases
As the proposition of the next-generation Web – semantic Web, semantic computing has been drawing more and more attention within the circle and the industries. A lot of research has been conducted on the theory and methodology of the subject, and potential applications have also been investigated and proposed in many fields. The progress of semantic computing made so far cannot be detached from...
متن کاملAn Association Network for Computing Semantic Relatedness
To judge how much a pair of words (or texts) are semantically related is a cognitive process. However, previous algorithms for computing semantic relatedness are largely based on co-occurrences within textual windows, and do not actively leverage cognitive human perceptions of relatedness. To bridge this perceptional gap, we propose to utilize free association as signals to capture such human p...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Neurocomputing
سال: 2021
ISSN: ['0925-2312', '1872-8286']
DOI: https://doi.org/10.1016/j.neucom.2020.09.075