Convolutional Neural Network Copy Detection with Neural Network Perceptual Hashing

In recent years, many model intellectual property (IP) proof methods for IP protection have been proposed, such as model watermarking and model fingerprinting. However, as an important part of the model IP protection system, the model copy detection task has not received enough attention. With the increasing number of neural network models transmitted and deployed on the Internet, the search for similar models is in great demand, which concurrently triggers the security problem of copy detection of models for IP protection. Due to the high computational complexity, both model watermarking and model fingerprinting lack the capability to efficiently find suspected infringing models among tens of millions of models. In this paper, inspired by the hash-based image retrieval methods, we propose a perceptual hashing algorithm for convolutional neural networks (CNNs). The proposed perceptual hashing algorithm can convert the weights of CNNs to fixed-length binary hash codes so that the lightly modified version has the similar hash code as the original model. By comparing the similarity of a pair of hash codes between a query model and a test model in the model library, the similar versions of a query model can be retrieved efficiently. To the best of our knowledge, this is the first perceptual hashing algorithm for CNNs. The experiment performed on a model library containing 3,565 models indicates that our proposed perceptual hashing scheme has a superior copy detection performance.

Convolutional Neural Network Copy Detection with Neural Network Perceptual Hashing

In recent years, many model intellectual property (IP) proof methods for IP protection have been proposed, such as model watermarking and model fingerprinting. However, as an important part of the model IP protection system, the model copy detection task has not received enough attention. With the increasing number of neural network models transmitted and deployed on the Internet, the search for similar models is in great demand, which concurrently triggers the security problem of copy detection of models for IP protection. Due to the high computational complexity, both model watermarking and model fingerprinting lack the capability to efficiently find suspected infringing models among tens of millions of models. In this paper, inspired by the hash-based image retrieval methods, we propose a perceptual hashing algorithm for convolutional neural networks (CNNs). The proposed perceptual hashing algorithm can convert the weights of CNNs to fixed-length binary hash codes so that the lightly modified version has the similar hash code as the original model. By comparing the similarity of a pair of hash codes between a query model and a test model in the model library, the similar versions of a query model can be retrieved efficiently. To the best of our knowledge, this is the first perceptual hashing algorithm for CNNs. The experiment performed on a model library containing 3,565 models indicates that our proposed perceptual hashing scheme has a superior copy detection performance.

Simulation and optimization of VRPSA system based on pseudo-transient method

In order to address the challenge of pressure swing adsorption system optimization, an optimization framework based on pseudo transient continuation method was used and vacuum rapid pressure swing adsorption process for oxygen production as a typical example. A pseudo transient model library was established and a robust two-stage dynamic tearing model was proposed to deal with the cyclic steady state conditions. Furthermore, the time constants were discussed and a practical time constant strategy and expressions were proposed for the stability and efficiency of calculation. Finally, reduced successive quadratic programming and time relaxation algorithm were used for the optimization of the two systems respectively, and the optimization results showed that although the simulation time of pseudo transient system is slightly higher than that of single discretization system, the optimization expense of single discretization system in two cases were 5.7 times and 11.6 times of that of pseudo transient system respectively.

Active participation of industrial enterprises in electricity markets - a generic modeling approach

Industrial enterprises represent a significant portion of electricity consumers with the potential of providing demand-side energy flexibility from their production processes and on-site energy assets. Methods are needed for the active and profitable participation of such enterprises in the electricity markets especially with variable prices, where the energy flexibility available in their manufacturing, utility and energy systems can be assessed and quantified. This paper presents a generic model library equipped with optimal control for energy flexibility purposes. The components in the model library represent the different technical units of an industrial enterprise on material, media, and energy flow levels with their process constraints. The paper also presents a case study simulation of a steel-powder manufacturing plant using the model library. Its energy flexibility was assessed when the plant procured its electrical energy at fixed and variable electricity prices. In the simulated case study, flexibility use at dynamic prices resulted in a 6% cost reduction compared to a fixed-price scenario, with battery storage and the manufacturing system making the largest contributions to flexibility.

Cloud-edge collaboration based transferring prediction of building energy consumption

Building energy consumption (BEC) prediction often requires constructing a corresponding model for each building based historical data. However, the constructed model for one building is difficult to be reused in other buildings. Recent approaches have shown that cloud-edge collaboration architecture is promising in realizing model reuse. How to complete the reuse of cloud energy consumption prediction models at the edge and reduce the computational cost of the model training is one of the key issues that need to be solved. To handle the above problems, a cloud-edge collaboration based transferring prediction method for BEC is proposed in this paper. Specifically, a model library stored prediction models for different types of buildings is constructed based the historical energy consumption data and the long short-term memory (LSTM) network in the cloud firstly; then, the similarity measurement strategies of time series with different granularity are given, and the model to be transferred from the model library is matched by analyzing the similarity between observation data uploaded to the cloud and the historical data collected in the cloud; finally, the fine-tuning strategy of the matching prediction model is given, and this model is fine-tuned at the edge to achieve its reuse in concrete application scenarios. Experiments on practical datasets reveal that compared with the prediction model which doesn’t utilize the transfer strategy, the proposed prediction model has better performance according to MAE and RMSE. Experimental results also confirm that the proposed method effectively reduces the computational cost of the network training at the edge.

Model Libraries: Historical Outline of the Project

The article presents the history of interaction between the government authorities and the library and information sector in the project of development of municipal libraries. The project resulted in the “model libraries”. The project has been operating since the early 2000s. The purpose of the article is to reveal the implementation and analyse the support at the level of federal and regional government management of the program until 2018 (the start of work within the frames of the National project “Culture”). The author presents periodization of the project, as well as quantitative characteristics and approaches to the understanding the concept of “model library”.Model library is a public library of municipal level that has standard and optimal set of material and information resources. At the same time, there is no approved definition of this phenomenon. It is assumed that the model rural library in its functions, content and equipment meets international and domestic standards and serves as a model for other institutions. The “model” indicators are reflected in the “Model Standard for Public Library Services”.The project on creation of model libraries on the basis of municipal public libraries in rural areas of Russia started in 2002 with the emergence of the all-Russian project “Creation of model public libraries in rural areas”. Since 2012, funding under the federal target program “Culture of Russia (2012—2018)” is carried out in the form of subsidies allocated on a competitive basis. The project became a “catalyst” for modernization processes in rural libraries of Russian regions. The transformation program included the following stages: repair and equipment of the library premises; acquisition of books and electronic publications; conducting training seminars for staff. The rural model library targeted the goals of ensuring equal and free access to information, creation of a comfortable library space, etc.After the approval of the “Model Standard for Public Library Services” (2014), it was decided to test its feasibility by creating so-called “pilot libraries of a new type”. According to this document, model libraries are intellectual centres equipped with high speed Internet, access to modern domestic information resources, which should provide access points to the National Electronic Library.The regional leaders in creation of model libraries are the Republic of Chuvashia, the Belgorod Region, the Republic of Bashkortostan, the Kursk Region and the Republic of Mari El. As of January 1, 2019, there were 3,310 model libraries operating in the country. The creation of model municipal libraries has become one of the departmental projects of the National project “Culture”.

Learning Strategies for E-Learning In Primary Schools Through Use of Zoom Meeting Technology in the Covid-19 Pandemic Era

During the Covid-19 pandemic, all learning activities were carried out through technology-based learning media, to avoid the spread of the Covid-19 virus. The purpose of this study was to find out how E-learning learning strategies in elementary schools through the use of Zoom Meeting technology in the era of the Covid-19 pandemic. The method in this research is to use the library research model (Library Research). The results showed that the learning media using Zoom Meeting technology as the most effective learning media used during learning in elementary schools, because it provides convenience in face-to-face interaction with teachers and peers. Keywords: Zoom Meeting, Learning Strategies, Learning Media