NOTICE OF RETRACTION: While investigating potential publication-related misconduct in connection with the ICIMTech 2021 Conference Proceedings, serious concerns were raised that cast doubt on the integrity of the peer-review process and all papers published in the Proceedings of this Conference. The integrity of the entire Conference has been called into question. As a result, of its investigation, ACM has decided to retract the Entire Conference Proceedings and all related papers from the ACM Digital Library.
The first person who put forwards the constructivism is J.Piaget,a Sweden psychologist.The constructivism considers that learning is a process learner builds up his interior psychological token initiatively,and learning needs enriching cognition and understanding completely through cooperation.Teachers are leaders and students are actors of learning.The teaching methods of constructivism include bracket model,anchor model,situation model and enter into teaching according to situation.Constructivism has guiding significance to teaching reform in universities and colleges in evidence.According to the theory of constructivism,we should implement teaching reform systematically,build up prepared courses according to the contents,foster students' consciousness of learning independent through formulating approaches and,develop new models of campus circumstance so as to promote healthy psychological development of students.
Practice training is an important part of the training program for engineering professionals in local agricultural universities.The monitoring and evaluation of teaching quality is very important.On the basis of a correct understanding of the basic problems of quality evaluation in the practical training link of engineering practice, it follows the principles of combining comprehensiveness and effectiveness, combining objective and operational, combining static and dynamic characteristics, and from the four major aspects of practice teaching management, base construction, content design and implementation, quality assessment and feedback.On the other hand, the quality index of each observation point is arranged reasonably, the corresponding quality evaluation standard is formulated, and a comprehensive and general practice training link quality evaluation system is constructed.
We have demonstrated a new type of transferred electron effect amplifier diode through computer simulation and experiment. The new GaAs diode is subcritically doped and has a cathode which limits the injection of carriers into the diode. In subcritically-doped amplifier diodes with ohmic contacts, excess carrier injection into the n-layer creates a power absorbing low field region near the cathode. Furthermore, the excess carrier injection causes the negative conductance to decrease rapidly with increasing rf drive and to disappear completely at a relatively low power level.
Time-series data typically contain issues such as missing data and noise, which can impact the model's precision and stability. This paper proposes a Transformer structure-based visual information-guided temporal data modeling algorithm to address the issues as mentioned above. The algorithm effectively captures the time-series structure of the time-series data, thereby enhancing the model's precision and stability. To evaluate the performance of the proposed algorithm, a dataset containing visual information aligned with time-series data is compiled, and a comprehensive quantitative and qualitative analysis is performed. Conduct a comprehensive quantitative and qualitative analysis. The results indicate that visual information can assist time-series data in capturing the intricate dynamics of the time-series data, thereby enhancing the performance of the proposed algorithm and facilitating its comprehension. The results indicate that visual information can assist time-series data in capturing the complex dynamics of time-series data, and thus in comprehending and predicting their behavior and trends. The application of this algorithm will advance research in the field of modeling and predicting time series data. Applying this algorithm will advance research and practice in modeling and forecasting time series data.
Secondary college management mode of graduate training in the university is established during the exploration of graduate training, which is the base of guarantee and improvement of the graduate teaching quality.The graduate training management mode in the secondary college is brought up as per the current situation of the graduate training and management in secondary college, such as shortage of recruitment of students, defective management system and not strong teacher team.The graduate training quality will be improved through the formulate and perfect graduate management system in the secondary college, the breakthrough of the recruitment of students, the construction of the teacher team and management, and double step management system etc..
Oil-immersed paper is a weak link in internal insulation of oil immersed power transformer, whose aging state distributes inhomogeneously. However, most of the current assessment methods fail to consider the spatial distribution of aging areas inside transformer. Aiming at this problem, this paper proposes the inversion detection method to obtain the resistivity of oil-immersed paper in various regions nondestructively, and uses the resistivity of oil-immersed paper to assess directly the insulation state of oil immersed power transformers. Finite element method is applied to establish the mapping relationship between the resistivity of oil-immersed paper in different regions, and the dielectric loss factor obtained from the port of transformer. Back propagation neural network is used to learn inversely this mapping relationship so as to calculate the resistivity. Applying the proposed method on a 10-kV transformer, the calculated results are close to the measured values. The study shows that the inversion method can effectively calculate the oil-immersed resistivity in various regions, which represent the aging state in different regions of the transformer.
Fine-grained classification poses significant challenges due to high intra-class variability and high inter-class similarity, making coarse-grained classification methods inadequate. Previous research has focused on locating objects in images while overlooking their detailed features. Additionally, images contain not only targets but also redundant information such as background and noise. The relationships between extracted features are also crucial, a point often overlooked in previous studies. To address these challenges, this paper proposes the Multi-scale Graph Neural Network Filter (MGF) model, composed of two modules: the Filter High-Resolution Feature Pyramid Network (FHRFPN) and the Adaptive Graph Neural Network (AGNN). FHRFPN progressively fuses features of different scales to retain rich detail information while avoiding conflicts between information and locating target positions. Before outputting, it employs the Channel Filter (CF) block to filter out noise features and reduce unnecessary computational overhead. AGNN adaptively creates adjacency matrices between features based on the input features. Through graph neural networks, the model comprehensively considers the interactions between features, effectively capturing non-linear and complex dependencies. The proposed method achieves state-of-the-art performance on the NA-Birds and CUB-200-2011 benchmarks, thus providing a promising solution for enhancing the performance of fine-grained visual classification tasks.
A new thin film pulse transformer for using in ISND and model systems is fabricated by a mask sputtering process. This novel pulse transformer consists of four I-shaped CoZrRe nanometer crystal magnetic-film cores and a Cu thin film coil, deposited on the micro-crystal glass substrate directly. The thickness of thin film core is between 1 and 3 μm, and the area is between 4mm×6 mm and 12mm×6 mm. The coils provide a relatively high induce of 0.8 μm and can be well operated in a frequency range of 0.001~20 MHz. A great interest has been attracted in planar magnetic devices to miniature various electronic equipment including pulse transformers and inductors, especially for the IT electronics applications and ISND modem systems, such as switching converters and inverters in portable equipments (1~3) . In Internet system, the thin film pulse transformer, with the sandwich structure of core/coil/core and coil/core/coil, in general, will be fabricated directly on ceramics substrate by lithography and ion etching dry process. The planar cores with zigzag and spiral coil fabricated on the substrate show high core loss and large magnetic-flux-leakage because their magnetic circuit is not close. Moreover, the lithography and ion etching dry process are very complex and need lots of advanced process systems, such as nitrogen ion beam etching system and sputtering system, thus the price of transformer is very high. In this work, a new pulse micro-transformer fabricated by mask vacuum sputtering process has been described. The pulse thin film transformer is composed of four I-shaped CoZrRe amorphous magnetic layers wound with Cu thin film coil, with primary and secondary coil turn of 3:3 and 3:2. The measured results show that the I-Shaped thin films transformer has a high Q value and good operation properties at high frequency, especially in range of 0.001~20 MHz. experiments is shown in Fig.1. The coils wound around the center of CoZrRe I-shaped magnetic core are connected to the upper and lower Cu patterns. The angle between the primary coil and the parallel direction is 22°, while as to the secondary coil this angle is 27°. In order to increase the mutual inductor, the primary and secondary coils are wound in an intercourse way. The length and width of the core are 6 mm and 3 mm, respectively. The coil width is 0.3 mm, and the width between adjacent coils is 0.5 mm apart.
The Graphics Processing Unit (GPU) has extended its applications from its original graphic rendering to more general scientific computation. Through massive parallelization, state-ofthe-art GPUs can deliver 200 billion floating-point operations per second (0.2 TFLOPS) on a single consumer-priced graphics card. This paper describes our attempt in leveraging GPUs for efficient HMM model training. We show that using GPUs for a specific example of Gaussian clustering, as required in fMPE, or feature-domain Minimum Phone Error discriminative training, can be highly desirable. The clustering of huge number of Gaussians is very time consuming due to the enormous model size in current LVCSR systems. Comparing an NVidia Geforce 8800 Ultra GPU against an Intel Pentium 4 implementation, we find that our brute-force GPU implementation is 14 times faster overall than a CPU implementation that uses approximate speed-up heuristics. GPU accelerated fMPE reduces the WER 6% relatively, compared to the maximumlikelihood trained baseline on two conversational-speech recognition tasks.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)