No abstract

We report dual ion beam sputtering fabrication of an Al/ZnO/Al memristor displaying forming-free bipolar resistive switching characteristics with memristive behavior without necessitating any post-processing steps. A nearly amorphous ZnO thin film and an appropriate concentration of oxygen vacancies play a significant role in imparting forming-free, stable, and reliable behavior to memory cells. Besides, sufficient non-lattice oxygen ions in the film play a crucial role in the resistive switching process. The AlOx interface layer is observed to strongly affect the switching mechanism in the memory device by altering the barrier at the Al/ZnO interface. The device shows stable switching behavior for >250 cycles with good retention and stable set/reset voltages.

Single synaptic device with inherent learning and memory functions is demonstrated based on a forming-free amorphous YO (yttria) memristor fabricated by dual ion beam sputtering system. Synaptic functions such as nonlinear transmission characteristics, long-term plasticity, short-term plasticity and 'learning behavior (LB)' are achieved using a single synaptic device based on cost-effective metal-insulator-semiconductor (MIS) structure. An 'LB' function is demonstrated, for the first time in the literature, for a yttria based memristor, which bears a resemblance to certain memory functions of biological systems. The realization of key synaptic functions in a cost-effective MIS structure would promote much cheaper synapse for artificial neural network.

No abstract

This survey paper focuses on one of the current primary issues challenging data mining researchers experimenting on real-world datasets. The problem is that of imbalanced class distribution that generates a bias toward the majority class due to insufficient training samples from the minority class. The current machine learning and deep learning algorithms are trained on datasets that are insufficiently represented in certain categories. On the other hand, some other classes have surplus samples due to the ready availability of data from these categories.Conventional solutions suggest undersampling of the majority class and/or oversampling of the minority class for balancing the class distribution prior to the learning phase. Though this problem of uneven class distribution is, by and large, ignored by researchers focusing on the learning technology, a need has now arisen for incorporating balance correction and data pruning procedures within the learning process itself. This paper surveys a plethora of conventional and recent techniques that address this issue through intelligent representations of samples from the majority and minority classes, that are given as input to the learning module. The application of nature-inspired evolutionary algorithms to intelligent sampling is examined, and so are hybrid sampling strategies that select and retain the difficult-to-learn samples and discard the easy-to-learn samples. The findings by various researchers are summarized to a logical end, and various possibilities and challenges for future directions in research are outlined. K E Y W O R D Sclass-imbalance problem, hybrid sampling, imbalanced data, oversampling, sampling, undersampling INTRODUCTIONLearning from imbalanced datasets results in a bias toward the majority class whose labeled samples are available in plenty as compared to the insufficiently represented minority class. 1 In data mining, factors that bring down the classifier performance are the intrinsic characteristics of the data and an uneven class distribution. 2 Lack of adequate data in the minority class results in a fuzzy and ever-varying decision boundary, leading to erroneous results. The problem isThis is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.