Value Of Loss Function Research Articles

Detection of clavicle fracture after martial arts training based on 3D segmentation and multi-perspective fusion

To improve the diagnosis efficiency of clavicle fractures, a Diagnosis Model of Clavicle Fracture Based on 3D Segmentation and Multi Perspective Fusion (3Ds MPF) is proposed in the experiment. By performing 3D segmentation on clavicle images, key layer images are extracted, and multi-perspective image data are fused and output to achieve accurate classification and diagnosis results. The results show that on the relevant dataset, the loss function value of the constructed model is less than 0.01, and the calculation accuracy is as high as 0.999. The average classification accuracy in actual use exceeds 90%, and it can capture the majority of fracture features in clavicle computed tomography images. The above results indicate that the model constructed by the research institute can significantly improve the judgment efficiency of computed tomography images, which helps to enhance the detection and diagnosis of clinical fracture conditions. The experimental results have helped improve the clinical diagnostic efficiency of fracture detection work.

Dynamical transition in controllable quantum neural networks with large depth

Understanding the training dynamics of quantum neural networks is a fundamental task in quantum information science with wide impact in physics, chemistry and machine learning. In this work, we show that the late-time training dynamics of quantum neural networks with a quadratic loss function can be described by the generalized Lotka-Volterra equations, leading to a transcritical bifurcation transition in the dynamics. When the targeted value of loss function crosses the minimum achievable value from above to below, the dynamics evolve from a frozen-kernel dynamics to a frozen-error dynamics, showing a duality between the quantum neural tangent kernel and the total error. In both regions, the convergence towards the fixed point is exponential, while at the critical point becomes polynomial. We provide a non-perturbative analytical theory to explain the transition via a restricted Haar ensemble at late time, when the output state approaches the steady state. Via mapping the Hessian to an effective Hamiltonian, we also identify a linearly vanishing gap at the transition point. Compared with the linear loss function, we show that a quadratic loss function within the frozen-error dynamics enables a speedup in the training convergence. The theory findings are verified experimentally on IBM quantum devices.

A Joint Estimation Method of Distribution Network Topology and Line Parameters Based on Power Flow Graph Convolutional Networks

Accurate identification of network topology and line parameters is essential for effective management of distribution systems. An innovative joint estimation method for distribution network topology and line parameters is presented, utilizing a power flow graph convolutional network (PFGCN). This approach addresses the limitations of traditional methods that rely on costly voltage phase angle measurements. The node correlation principle is applied to construct a node correlation matrix, and a minimum distance iteration algorithm is proposed to generate candidate topologies, which serve as graph inputs for the parameter estimation model. Based on the topological dependencies and convolutional properties of AC power flow equations, a PFGCN model is designed for line parameter estimation. Parameter refinement is achieved through an alternating iterative process of pseudo-trend calculation and neural network training. Training convergence and loss function values are used as feedback to filter and validate candidate topologies, enabling precise joint estimation of both topologies and parameters. The proposed method’s accuracy, transferability, and robustness are demonstrated through experiments on the IEEE-33 and modified IEEE-69 distribution systems. Multiple metrics, including MAPE, IAE, MAE, and R2, highlight the proposed method’s advantages over Adaptive Ridge Regression (ARR). In the C33 scenario, the proposed method achieves MAPEs of 4.6% for g and 5.7% for b, outperforming the ARR method with MAPEs of 7.1% and 7.9%, respectively. Similarly, in the IC69 scenario, the proposed method records MAPEs of 3.0% for g and 5.9% for b, surpassing the ARR method’s 5.1% and 8.3%.

Study on long short-term memory based on vector direction of flood process for flood forecasting

Accurate flood forecasting is crucial for flood prevention and mitigation, safeguarding the lives and properties of residents, as well as the rational use of water resources. The study proposes a model of long and short-term memory (LSTM) combined with the vector direction (VD) of the flood process. The Jingle and Lushi basins were selected as the research objects, and the model was trained and validated using 50 and 49 measured flood rainfall-runoff data in a 7:3 division ratio, respectively. The results indicate that the VD-LSTM model has more advantages than the LSTM model, with increased NSE, and reduced RMSE and bias to varying degrees. The flow simulation results of VD-LSTM better match the observed flow hydrographs, improving the underestimation of peak flows and the lag issue of the model. Under the same task and dataset, with the same hyperparameter settings, VD-LSTM can more quickly reduce the loss function value and achieve a better fit compared to LSTM. The proposed VD-LSTM model couples the vectorization process of flood runoff with the LSTM neural network, which contributes to the model better exploring the change characteristics of rising and receding water in flood runoff processes, reducing the training gradient error of input–output data for the LSTM model, and more effectively simulating flood process.

Quantitative strategies of different loss functions aggregation for knowledge distillation

<p>Deep learning models have been successfully applied to many visual tasks. However, they tend to be increasingly cumbersome due to their high computational complexity and large storage requirements. How to compress convolutional neural network (CNN) models while still maintain their efficiency has received increasing attention from the community, and knowledge distillation (KD) is efficient way to do this. Existing KD methods have focused on the selection of good teachers from multiple teachers, or KD layers, which is cumbersome, expensive computationally, and requires large neural networks for individual models. Most of teacher and student modules are CNN-based networks. In addition, recent proposed KD methods have utilized cross entropy (CE) loss function at student network and KD network. This research focuses on the quantifiable evaluation of teacher-student model, in which knowledge is not only distilled from training models that have the same CNN architecture but also from different architectures. Furthermore, we propose combination of CE, balance cross entropy (BCE), and focal loss functions to not only soften the value of loss function in transferring knowledge from large teacher model to small student model but also increase classification performance. The proposed solution is evaluated on four benchmark static image datasets, and the experimental results show that our proposed solution outperforms the state-of-the-art (SOTA) methods from 2.67% to 9.84% at top 1 accuracy.</p>

Credit risk prediction for small and micro enterprises based on federated transfer learning frozen network parameters

To accelerate the convergence speed and improve the accuracy of the federated shared model, this paper proposes a Federated Transfer Learning method based on frozen network parameters. The article sets up frozen two, three, and four layers network parameters, 8 sets of experimental tasks, and two target users for comparative experiments on frozen network parameters, and uses homomorphic encryption based Federated Transfer Learning to achieve secret transfer of parameters, and the accuracy, convergence speed, and loss function values of the experiment were compared and analyzed. The experiment proved that the frozen three-layer network parameter model has the highest accuracy, with the average values of the two target users being 0.9165 and 0.9164; The convergence speed is also the most ideal, with fast convergence completed after 25 iterations. The training time for the two users is also the shortest, with 1732.0s and 1787.3s, respectively; The loss function value shows that the lowest value for User-II is 0.181, while User-III is 0.2061. Finally, the unlabeled and non-empty enterprise credit data is predicted, with 61.08% of users being low-risk users. This article achieves rapid convergence of the target network model by freezing source domain network parameters in a shared network, saving computational resources.

Image acquisition technology for unmanned aerial vehicles based on YOLO - Illustrated by the case of wind turbine blade inspection

Wind energy, as a renewable energy source, is becoming increasingly important. The maintenance and damage detection of wind turbine blades are particularly crucial. For this purpose, the study aims to optimize the You Only Look Once (YOLO) processing algorithm for drone images to improve the detection efficiency. Firstly, the damage images captured by drones are preprocessed and optimized, including deblurring, noise reduction, and image enhancement. Subsequently, the YOLOv5 model is improved in terms of structure and regression function, and a novel damage detection model is proposed. The research results indicated that the minimum loss function value of the improved model was 2.75, the average accuracy was 95 %, and the highest intersection over union was 91 %. After simulation testing, the detection effect of this model on abrasion, crackle, edge cracking, and coating peeling images was significantly better than other models in the same series. Its average time was as short as 2.43 s, reaching a maximum frame rate of 35.46. From this, the combination of drone image technology and improved image processing algorithm has a positive impact on improving the operational efficiency and safety of wind turbine blades. Compared with the traditional methods, the proposed model has significant advantages in terms of accuracy and real-time performance of damage detection, providing a new technical means for efficient maintenance of wind turbines. Meanwhile, the method shows high robustness and reliability in different types of damage detection, demonstrating the extensive potential in practical applications.

Anomaly detection combining bidirectional gated recurrent unit and autoencoder in the context of e-commerce

E-commerce platforms store a large amount of user personal information, transaction data, and financial information, which have extremely high value for hackers and criminals. Therefore, protecting the security of e-commerce platforms is particularly important, and intrusion detection is a technical means used to discover and respond to possible security threats and attacks. But with the development of Internet technology, there are more and more types of intrusion attacks and more sophisticated means. Traditional intrusion detection systems are difficult to cope with. This study proposes an anomaly detection model based on bidirectional gated loop units and autoencoders. The model learns HTTP text data, trains the model, and uses bidirectional gated loop units to convert text sequences from characters to numbers. The experimental results show that when the training set size is 1000, the false alarm rates of Analytic Hierarchy Process, Support Vector Machine, Long Short Term Recurrent Memory Network, and Improved end-to-end algorithm models are 0.30, 0.27, 0.23, and 0.10, respectively. The loss function values are 0.35, 0.28, 0.17, and 0.13, respectively. The F1 values are 0.78, 0.88, 0.91, and 0.99, and the accuracy rates are 0.88, 0.91, 0.95, and 0.99, respectively. The research results indicate that the proposed method model has excellent performance.

Pressure-dependent band gap engineering with structural, electronic, mechanical, optical, and thermal properties of CsPbBr3: first-principles calculations.

In the renewable industry, pressure-dependent CsPbBr3 perovskite has a lot of potential due to its exceptional properties. Present work revealed the mechanical stability of CsPbBr3 between 0 to 50GPa. The bandgap of unstressed CsPbBr3 is 2.90eV, indicating a direct bandgap. Band gap values decrease by increasing external pressure. CsPbBr3 structure showed a direct band gap from 0 to 35GPa and in-direct from 40 to 50GPa. The unit cell volume and lattice constants are substantially decreased. Mechanical parameters, i.e., Young's modulus, bulk modulus, anisotropy factor, shear modulus, and poison's ratio are obtained. Under ambient conditions, the mechanical properties of CsPbBr3 showed ductile behavior and with induced pressure, their ductility has significantly improved. By applying stresses ranging from 0 to 50GPa, the considerable fluctuation in values of dielectric function (imaginary and real), absorption, reflectivity, loss function, refractive index (imaginary and real), and conductivity (imaginary and real), was also identified. When pressure rises, the optical parameters increase and drag in the direction of high energies. Response functions are used to predict the density of states and the phonon lattice dispersion to study the phonon properties. By using the quasi-harmonic Debye model, the thermal effect on the free energy, entropy, enthalpy, and heat capacity were predicted and compared. These results would be useful for theoretical research and indicate how external pressure significantly affects the physical characteristics of CsPbBr3 perovskites, which may open up new possibilities for use in optoelectronic, photonic, and solar cell applications. The structural, electrical, mechanical, optical, and thermal properties of cesium lead bromide (CsPbBr3) are investigated by applying external pressure from 0 to 50GPa, using generalized gradient approximations (GGA) and Perdew-Burke-Ernzerhof (PBE) with CASTEP code built-in material studio by density functional theory (DFT).

Atmospheric Turbulence Phase Reconstruction via Deep Learning Wavefront Sensing.

The fast and accurate reconstruction of the turbulence phase is crucial for compensating atmospheric disturbances in free-space coherent optical communication. Traditional methods suffer from slow convergence and inadequate phase reconstruction accuracy. This paper introduces a deep learning-based approach for atmospheric turbulence phase reconstruction, utilizing light intensity images affected by turbulence as the basis for feature extraction. The method employs extensive light intensity-phase samples across varying turbulence intensities for training, enabling phase reconstruction from light intensity images. The trained U-Net model reconstructs phases for strong, medium, and weak turbulence with an average processing time of 0.14 s. Simulation outcomes indicate an average loss function value of 0.00027 post-convergence, with a mean squared error of 0.0003 for individual turbulence reconstructions. Experimental validation yields a mean square error of 0.0007 for single turbulence reconstruction. The proposed method demonstrates rapid convergence, robust performance, and strong generalization, offering a novel solution for atmospheric disturbance correction in free-space coherent optical communication.

Practical one-shot data-driven design of fractional-order PID controller: Fictitious reference signal approach

This study proposes a one-shot data-driven tuning method for a fractional-order proportional-integral-derivative (FOPID) controller. The proposed method tunes the FOPID controller in the model-reference control formulation. A loss function is defined to evaluate the match between a given reference model and the closed-loop response while explicitly considering the closed-loop stability. A loss function value is based on the fictitious reference signal computed using the input/output data. Model matching is achieved via loss function minimization. The proposed method is simple and practical: it needs only one-shot input/output data of a plant (no plant model required), considers the bounded-input bounded-output stability of the closed-loop system from a bounded reference input to a bounded output, and automatically determines the appropriate parameter value via optimization. Numerical simulations show that the proposed approach facilitates good control performance, and destabilization can be avoided even if perfect model matching is unachievable.

Three-dimensional artistic design method of ceramic products based on recurrent neural network technology

Three-dimensional digital technology has made breakthroughs and shown unique advantages in all walks of life. On the basis of practicality, the three-dimensional artistic design of ceramic products gradually adds some aesthetic, artistic design elements, which brings beautiful enjoyment to people’s lives and makes people’s lives colorful. This paper presents a three-dimensional artistic design method for ceramic products based on RNN (recurrent neural network) technology. With the establishment of the 3D YOLOv3 framework, the new model training is faster and more stable, the convergence speed of the loss function is faster, and the reconstructed 3D model is more accurate. After training for a certain number of times, the network gradually becomes stable, the accuracy rate is kept at 95%, and the loss function value is reduced below 0.2. The accuracy of the network model and the precision of semantic segmentation are improved. The semantic segmentation and object recognition under 3D scene reconstruction studied in this paper have certain theoretical value and high feasibility.

On the practical usefulness of the Hardware Efficient Ansatz

Variational Quantum Algorithms (VQAs) and Quantum Machine Learning (QML) models train a parametrized quantum circuit to solve a given learning task. The success of these algorithms greatly hinges on appropriately choosing an ansatz for the quantum circuit. Perhaps one of the most famous ansatzes is the one-dimensional layered Hardware Efficient Ansatz (HEA), which seeks to minimize the effect of hardware noise by using native gates and connectives. The use of this HEA has generated a certain ambivalence arising from the fact that while it suffers from barren plateaus at long depths, it can also avoid them at shallow ones. In this work, we attempt to determine whether one should, or should not, use a HEA. We rigorously identify scenarios where shallow HEAs should likely be avoided (e.g., VQA or QML tasks with data satisfying a volume law of entanglement). More importantly, we identify a Goldilocks scenario where shallow HEAs could achieve a quantum speedup: QML tasks with data satisfying an area law of entanglement. We provide examples for such scenario (such as Gaussian diagonal ensemble random Hamiltonian discrimination), and we show that in these cases a shallow HEA is always trainable and that there exists an anti-concentration of loss function values. Our work highlights the crucial role that input states play in the trainability of a parametrized quantum circuit, a phenomenon that is verified in our numerics.

Wind direction and speed calculation model with deep learning

Abstract In view of the problems of complex process and weak representation in regular atmospheric wind field detection, this paper adopts deep learning method, uses global high-altitude meteorological detection data, and establishes a deep learning model for calculating wind direction and speed with different altitudes and temperatures by using keras software package. The model is verified by using third-party independent sounding data from a meteorological observatory in Shanghai. The calculation accuracy of the model above 2000 m is 0.9830 and the value of loss function is 0.0482. The accuracy under 2000 m is 0.9164 and the value of loss function is 0.0377. There are significant differences in the performance of the model between under 2000 meters and above 2000 meters due to surface friction. The model shows that wind direction and speed of different height layers can be calculated by using only height and temperature at the same height. This model can also be used to check whether the quality of regular wind detection work is good or not with big old data.

DSE-NN: Discretized Spatial Encoding Neural Network for Ocean Temperature and Salinity Interpolation in the North Atlantic

The precise interpolation of oceanic temperature and salinity is crucial for comprehending the dynamics of marine systems and the implications of global climate change. Prior neural network-based interpolation methods face constraints related to their capacity to delineate the intricate spatio-temporal patterns that are intrinsic to ocean data. This research presents an innovative approach, known as the Discretized Spatial Encoding Neural Network (DSE-NN), comprising an encoder–decoder model designed on the basis of deep supervision, network visualization, and hyperparameter optimization. Through the discretization of input latitude and longitude data into specialized vectors, the DSE-NN adeptly captures temporal trends and augments the precision of reconstruction, concurrently addressing the complexity and fragmentation characteristic of oceanic data sets. Employing the North Atlantic as a case study, this investigation shows that the DSE-NN presents enhanced interpolation accuracy in comparison with a traditional neural network. The outcomes demonstrate its quicker convergence and lower loss function values, as well as the ability of the model to reflect the spatial and temporal distribution characteristics and physical laws of temperature and salinity. This research emphasizes the potential of the DSE-NN in providing a robust tool for three-dimensional ocean temperature and salinity reconstruction.

Energy Saving and Emission Reduction Optimization of Enterprise Hazardous Waste Recycling Management System based on Hybrid Genetic Algorithm

This research proposes a new path planning model for hazardous waste recycling transportation to effectively manage the hazardous waste recycling transportation, improving the transportation efficiency, while considering the actual road conditions. The new model adopts the conservation algorithm and hybrid genetic algorithm, which makes the new model better meet the complex needs of hazardous waste recycling. The new approach enables optimal transportation path planning for hazardous waste recycling while ensuring safety and compliance. The results showed that the hybrid algorithm outperformed the other two algorithms in terms of path optimization, cost reduction, accuracy improvement and error reduction. The hybrid algorithm had the best path optimization effect, which can get the optimal path with the lowest cost and highest efficiency. The hybrid algorithm had the highest accuracy of 95.62%. It also had the lowest root mean square error and average percentage error, indicating that it had less error. Finally, the hybrid algorithm had the highest loss function value, which indicated that the model had the best stability and better performance. The new hybrid genetic algorithm performed better than the single traditional algorithm, which is more efficient for hazardous waste recycling.

Automatic grid topology detection method based on Lasso algorithm and t-SNE algorithm

For a long time, the low-voltage distribution network has the problems of untimely management and complex and frequently changing lines, which makes the problem of missing grid topology information increasingly serious. This study proposes an automatic grid topology detection model based on lasso algorithm and t-distributed random neighbor embedding algorithm. The model identifies the household-variable relationship through the lasso algorithm, and then identifies the grid topology of the station area through the t-distributed random neighbor embedding algorithm model. The experimental results indicated that the lasso algorithm, the constant least squares algorithm and the ridge regression algorithm had accuracies of 0.88, 0.80, and 0.71 and loss function values of 0.14, 0.20, and 0.25 for dataset sizes up to 500. Comparing the time spent on identifying household changes in different regions, in Region 1, the training time for the Lasso algorithm, the Constant Least Squares algorithm, and the Ridge Regression algorithm is 2.8 s, 3.0 s, and 3.1 s, respectively. The training time in region 2 is 2.4s, 3.6s, and 3.4s, respectively. The training time in region 3 is 7.7 s, 1.9 s, and 2.8 s, respectively. The training time in region 4 is 3.1 s, 3.6 s, and 3.3 s, respectively. The findings demonstrate that the suggested algorithmic model performs better than the other and can identify the structure of LV distribution networks.

EPB-TBM cutterhead torque and thrust modelling in rock tunnels through an analytical method and TSFS model

The analysis of cutterhead torque and thrust force is of utmost importance in the selection of a proper tunnel boring machine (TBM) and in achieving optimal operation of the earth pressure balance (EPB) machine for a specific project. This study focuses on the analysis of cutterhead torque and thrust force using data collected from the recent metro project of Tehran, specifically the southern extension of Tehran metro line 6, (TML6-SE). To assess the impact of geotechnical and operational parameters on the measured torque and thrust, a database consisting of 457 data points was established. The excavation of a significant portion of the project took place in a dense and robust limestone using a refurbished EPB machine with a diameter of 9.19 m, initially designed for soft ground excavation. To complete this section of the tunnel, the cutterhead was equipped with disc cutters instead of rippers. Consequently, existing theoretical models for estimating torque and thrust of an EPB machine were modified to accommodate the hard rock boring process. Additionally, the Takagi-Sugeno fuzzy system (TSFS) was utilized in this study. The results of the models yielded quantitative conclusions, and the comparison between the measured and predicted target parameters demonstrated high accuracy. This study contributes to the understanding of the influence of various parameters on the performance of EPB machines in hard rock boring conditions, as well as providing insight into the potential adjustments needed in theoretical models for predicting cutterhead torque and thrust force. The proposed models showed satisfactory and reliable accuracy, with all models having an R2 value above 0.7 and loss function values nearing zero.

Multiplex Asymmetric PCR by Combining the Amplification Refractory Mutation System with the Homo-Tag-Assisted Nondimer System.

Asymmetric PCR is widely used to produce single-stranded amplicons (ss-amplicons) for various downstream applications. However, conventional asymmetric PCR schemes are susceptible to events that affect primer availability, which can be exacerbated by multiplex amplification. In this study, a new multiplex asymmetric PCR approach that combines the amplification refractory mutation system (ARMS) with the homo-Tag-assisted nondimer system (HANDS) is described. ARMS-HANDS (A-H) PCR utilizes equimolar-tailed forward and reverse primers and an excess Tag primer. The tailed primer pairs initiate exponential symmetric amplification, whereas the Tag primer drives linear asymmetric amplification along fully matched strands but not one-nucleotide mismatched strands, thereby generating excess ss-amplicons. The production of ss-amplicons is validated using agarose gel electrophoresis, sequencing, and melting curve analysis. Primer dimer alleviation is confirmed by both the reduced Loss function value and a 20-fold higher sensitivity in an 11-plex A-H PCR assay than in an 11-plex conventional asymmetric PCR assay. Moreover, A-H PCR demonstrates unbiased amplification by its allele quantitative ability in correct identification of all 31 trisomy 21 samples among 342 clinical samples. A-H PCR is a new generation of multiplex asymmetric amplification approach with various applications, especially when sensitive and quantitative detection is required.

Effects of different fire slash artificial promotion regeneration and natural material regeneration on ecological function

IntroductionIn the aftermath of a fire, prompt reforestation of the affected areas is crucial to mitigate economic losses and ecological impacts.MethodsThis paper introduces an ecological function assessment model leveraging the Back Propagation Neural Network (BPNN). The model's efficacy is validated through simulation comparison experiments. Subsequently, an analysis of the ecosystem's material circulation and energy flow capabilities is undertaken.ResultsSimulation outcomes reveal that our proposed model attains convergence by the 10th training iteration, with a loss function value of just 0.28, highlighting minimal training loss. This underscores the model's rapid convergence and impressive training performance. Our method proves superior to the comparison method in both initial and later operational phases. Notably, it offers a significantly faster response speed and boasts an accuracy rate exceeding 95%.DiscussionConsequently, employing this model to analyze ecological function changes is deemed feasible. The analysis of ecosystem material circulation and energy flow capabilities reveals that while initial assessments show minimal change, scores exhibit a clear acceleration as the cycle progresses.