We develop and explore deep learning single-shot ptychography. The deep learning algorithm, trained using only experimental data and without any model of the system, leads to significantly better reconstructions than ...
详细信息
Noise is a major concern in seismic data and influences the processing and interpretability of seismic data at various steps. However, noise has a certain pattern, which can be exploited by machine learning algorithms...
详细信息
The paper proposes a new method for compensation of the transmitter (TX) power amplifier (PA) nonlinearity and I/Q imbalance impairments at the receiver (RX) side for single carrier (SC) communication systems. This me...
详细信息
The identification of words in texts and speech is an important ingredient in speech and language recognition systems. Unsupervised learning algorithms use distributional information in texts to derive regularities th...
详细信息
The identification of words in texts and speech is an important ingredient in speech and language recognition systems. Unsupervised learning algorithms use distributional information in texts to derive regularities that the human brain would construe as lexical units, i.e. morphemes. Since statistical distributions of alphabetic or phonemic clusters are not knowingly experienced by the human mind, the exploitation of such information by a machine and making it accessible to our senses, results in new insights of the input material. The study at hand focuses on the properties of language input, which is systematically varied. It will be shown that the language register will not have an effect in English on the identification of words. Yet, there are significant differences if the form of representation is changed from an alphabetic to a phonemic representation. The control language, Japanese, reveals that it is not a universal feature among the languages of the world. Hence the design of algorithms exploiting distributional cues should be defined language-specifically. Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
The state grid information system is complex, the operation and maintenance information are diverse, involving a wide range of aspects. It becomes a key problem that how to use the alarm log of the operation and maint...
详细信息
Sensor fusion is the process of acquiring and fusing data from two different sensors for environment perception. With the increase in the level of automation needed for almost all the systems, depth perception has bec...
详细信息
Breast cancer is the most common malignant tumor among women, and the incidence is on the rise all over the world, endangering women's life and health seriously, the prevention and treatment of this disease become...
详细信息
Plant disease is one of the important threat factors that hinder the normal growth and development of plants. The intelligent identification of plant disease species has become increasingly important in the agricultur...
详细信息
The problem of network intrusion detection poses innumerable challenges to the research community, industry, and commercial sectors. Moreover, the persistent attacks occurring on the cyber-threat landscape compel rese...
详细信息
The problem of network intrusion detection poses innumerable challenges to the research community, industry, and commercial sectors. Moreover, the persistent attacks occurring on the cyber-threat landscape compel researchers to devise robust approaches in order to address the recurring problem. Given the presence of massive network traffic, conventional machine learning algorithms when applied in the field of network intrusion detection are quite ineffective. Instead, a hybrid multimodel solution when sought improves performance thereby producing reliable predictions. Therefore, this article presents an ensemble model using metaclassification approach enabled by stacked generalization. Two contemporary as well as heterogeneous datasets, namely, UNSW NB-15, a packet-based dataset, and UGR'16, a flow-based dataset, that were captured in emulated as well as real network traffic environment, respectively, were used for experimentation. Empirical results indicate that the proposed stacking ensemble is capable of generating superior predictions with respect to a real-time dataset (97% accuracy) than an emulated one (94% accuracy).
Reinforcement learning algorithms have been shown to converge to the classic replicator dynamics of evolutionary game theory, which describe the evolutionary process in the limit of an infinite population. However, it...
详细信息
暂无评论