Rate Adaptation Research Articles

Improving performance of WSNs in IoT applications by transmission power control and adaptive learning rates in reinforcement learning

Improving performance of WSNs in IoT applications by transmission power control and adaptive learning rates in reinforcement learning

Anterior cingulate cortex provides the neural substrates for feedback-driven iteration of decision and value representation

Adjusting decision-making under uncertain and dynamic situations is the hallmark of intelligence. It requires a system capable of converting feedback information to renew the internal value. The anterior cingulate cortex (ACC) involves in error and reward events that prompt switching or maintenance of current decision strategies. However, it is unclear whether and how the changes of stimulus-action mapping during behavioral adaptation are encoded, nor how such computation drives decision adaptation. Here, we tracked ACC activity in male mice performing go/no-go auditory discrimination tasks with manipulated stimulus-reward contingencies. Individual ACC neurons integrate the outcome information to the value representation in the next-run trials. Dynamic recruitment of them determines the learning rate of error-guided value iteration and decision adaptation, forming a non-linear feedback-driven updating system to secure the appropriate decision switch. Optogenetically suppressing ACC significantly slowed down feedback-driven decision switching without interfering with the execution of the established strategy.

A deep learning-based U-Net 3+ technique for segmentation blood cell

This study introduces a novel and enhanced UNet3Plus model tailored for the precise segmentation of blood cells in medical images. The architecture incorporates structural modifications, including strengthened connections between convolutional layers, increased filter numbers, and integration of Bayesian optimization for hyperparameter tuning. The model's generalization capability is optimized through the dynamic adjustment of dropout rates and learning rates. Bayesian optimization facilitates the exploration of optimal hyperparameter combinations, allowing the model to adapt effectively to diverse datasets. Advanced training strategies, such as adaptive learning rate adjustment and early stopping, are employed to mitigate overfitting and enhance training efficiency. The proposed model exhibits exceptional performance across multiple folds, achieving low training and validation losses, high accuracy metrics, and robust segmentation indices. Evaluation metrics, including Mean IoU (Jaccard Index), Dice score, Pixel Accuracy, and Precision, affirm the model's proficiency in accurately delineating blood cell boundaries. The study contributes to the field of deep learning-based medical image segmentation by showcasing the effectiveness of customized architectures and optimization techniques. The proposed UNet3Plus model stands as a promising solution for accurate and reliable blood cell segmentation, demonstrating adaptability and robust performance across various datasets. This work sets the stage for future research in the domain of medical image segmentation, emphasizing the potential for continued advancements in precise and efficient segmentation methodologies.

Control of a phytopathogenic infection in an apple orchard in the forest-steppe of Samara Oblast

The study evaluates a protection system by selecting and alternating fungicides having different mechanisms of action to exclude the emergence of resistant populations of apple tree pathogens. The experiments were started at Koshelevsky Posad LLC (Syzransky District, Samara Oblast) in 2022-2023. The orchard was planted in 2013-2014, with trees arranged in a 3.5×1.5 m grid; the 62-396 rootstock was used. The biological effectiveness of fungicide protection against apple scab (Venturia inaequalis (Cooke) Wint.) and fruit rot (Monilia fructigena Pers.) was analyzed using ‘Berkutovskoye,’ ‘Kulikovskoye,’ ‘Kuibyshevskoye,’ ‘Lobo,’ and ‘Northern Sinap’ varieties. The protection system including contact and systemic agents having different mechanisms of action was developed according to the FRAC diagram for assessing the combined risk of fungicide resistance developing in phytopathogens, which takes into account the adaptation rate of organisms (emergence of new resistant types) to changing environmental conditions. An analysis of the correlation between the growth rate of apple-specific phytopathogens and weather conditions shows no significant relationship (0.032-0.245). In these studies, the anthropogenic factor-chemical treatment with fungicides – had a predominant effect on the development of scab and monilia in apple orchards along with weather factors. The yield of marketable apples in the experimental quarter of the orchard differed insignificantly from year to year. In 2023, the yield was lower; however, this fact can be probably attributed to the return of late frosts during apple blossom in Samara Oblast. Noteworthy is that for the ‘Berkutovskoye’ and ‘Lobo’ varieties, unmarketable apples did not exceed 5 and 12 %, respectively, in the studied quarter of the orchard. The protection system using fungicides (GC Shans) helped to optimize the phytosanitary condition of the orchard, thus reducing the amount of fallen fruit and providing a stable harvest of marketable apples.

Advancing crystal growth prediction: An adaptive kMC model spanning multiple regimes

Traditional batch crystallization processes encounter challenges like batch-to-batch variability, and difficulty in scale-up and achieving desired crystal size and morphology. These challenges stem from various growth mechanisms within crystallizers, influenced by operating conditions that dictate the dynamic surface structure of crystals. To overcome this, we developed a novel microscopic kinetic Monte Carlo model, which uses a specialized adsorption rate that adapts to different growth regimes by considering surface thermodynamic effects for different growth mechanisms. Specifically, the adaptive adsorption rate redefines the driving force of crystallization by considering the attachment energies of different adsorption sites (e.g., kink, adatom, and edge) and supersaturation. This approach enables seamless transitions across growth regimes. We use the lysozyme crystal system as a case study, demonstrating the model's superior predictive power in regime-to-regime transitions. Overall, the developed work provides a valuable new approach to understanding crystal morphology and predicting crystal growth rates in different growth regimes.

Adaptive Segmented Aggregation and Rate Assignment Techniques for Flexible-Length Polar Codes.

Polar codes have garnered a lot of attention from the scientific community, owing to their low-complexity implementation and provable capacity achieving capability. They have been standardized to be used for encoding information on the control channels in 5G wireless networks due to their robustness for short codeword lengths. The conventional approach to generate polar codes is to recursively use 2×2 kernels and polarize channel capacities. This approach however, has a limitation of only having the ability to generate codewords of length Norig=2n form. In order to mitigate this limitation, multiple techniques have been developed, e.g., polarization kernels of larger sizes, multi-kernel polar codes, and downsizing techniques like puncturing or shortening. However, the availability of so many design options and parameters, in turn makes the choice of design parameters quite challenging. In this paper, the authors propose a novel polar code construction technique called Adaptive Segmented Aggregation which generates polar codewords of any arbitrary codeword length. This approach involves dividing the entire codeword into smaller segments that can be independently encoded and decoded, thereby aggregated for channel processing. Additionally a rate assignment methodology has been derived for the proposed technique, that is tuned to the design requirement.

Bucketized Active Sampling for learning ACOPF

This paper considers optimization proxies for Optimal Power Flow (OPF), i.e., machine-learning models that approximate the input/output relationship of OPF. Recent work has focused on showing that such proxies can be of high fidelity. However, their training requires significant data, each instance necessitating the (offline) solving of an OPF. To meet the requirements of market-clearing applications, this paper proposes Bucketized Active Sampling (BAS), a novel active learning framework that aims at training the best possible OPF proxy within a time limit. BAS partitions the input domain into buckets and uses an acquisition function to determine where to sample next. By applying the same partitioning to the validation set, BAS leverages labeled validation samples in the selection of unlabeled samples. BAS also relies on an adaptive learning rate that increases and decreases over time. Experimental results demonstrate the benefits of BAS.

Does Rate Adaptation at Daily Timescales Make Sense?

Today, networking hardware is not fast enough to save energy with rate adaptation. Or is it? While we are not (yet) able to turn on and off line cards in milliseconds, we can do so a couple of times per day. The question is, does it make sense to save energy? It may: We estimate we could potentially save in the order of MWh per year with link sleeping and down-rating in a cloud provider network. Importantly, these are "easy" gains: They would come without modifying the routing state nor impacting the quality of service of the traffic; it only leverages the typically low utilization of the links. This study provides only a rough approximation of the potential savings of rate adaptation and leaves several practical questions open. More than anything else, it motivates further investigation of the approach.

Evolution of "invasion syndrome" in invasive goldenrod is not constrained by genetic trade-offs.

A suite of plant traits is thought to make weed populations highly invasive, including vigorous growth and reproduction, superior competitive ability, and high dispersal ability. Using a breeding design and a common garden experiment, we tested whether such an "invasion syndrome" has evolved in an invasive range of Solidago altissima, and whether the evolution is likely to be genetically constrained. We found an overall shift in invasive phenotypes between native North American and invasive Japanese populations. The invasive populations were taller and produced more leaves, suggesting a superior ability to exploit limited resources. The populations also produced more allelopathic compounds that can suppress competitor growth. Finally, invasive populations produced more seeds, which are smaller and are released from a greater height, indicating a potential for superior dispersal ability than the native populations. Quantitative genetics analyses found a large amount of additive genetic variation in most focal traits across native and invasive populations, with no systematic differences in its magnitude between the ranges. Genetic covariances among three traits representing invasion strategies (leaf mass, polyacetylene concentration and seed size) were small. The R metric, which measures the effect of genetic covariances on the rate of adaptation, indicated that the covariance neither constrains nor accelerates concerted evolution of these traits. The results suggest that the invasion syndrome in S. altissima has evolved in the novel range due to ample additive genetic variation, and relatively free from genetic trade-offs.

Potato seed production of G2 from plantlet cutting by setting of population, spacing, and mulch

Abstract Quick multiplication of potato seeds using plantlet cuttings outside of screen house can accelerate the availability of high quality seeds, especially G2 seeds. However, some obstacles are faced, especially the low adaptation rate of plantlet cuttings. This study aimed to increase the production of potato seed G2 from plantlet cutting by using various population number per hole, spacing, and plastic mulch. The study used a randomized completed block design with three treatment factors and six replications. The first factor was the population number (1, 2, 3, and 4 cuttings per hole). The second was spacing in row (20 and 30 cm). The third was the usage of mulch (yes and no). The results showed that the technology combination of four cuttings per hole, spacing in rows of 30 cm, and mulch had the largest number of total and medium tubers per plant. The population at a planting distance of 30 cm which was reduced in this treatment could be increased by increasing the number of cuttings per hole, so that the total number of tubers remained high, especially when using mulch. Single cutting in no mulch and wider spacing would promote much number of large tubers.

Neural Network Signal Integration from Thermogas-Dynamic Parameter Sensors for Helicopters Turboshaft Engines at Flight Operation Conditions.

The article's main provisions are the development and application of a neural network method for helicopter turboshaft engine thermogas-dynamic parameter integrating signals. This allows you to effectively correct sensor data in real time, ensuring high accuracy and reliability of readings. A neural network has been developed that integrates closed loops for the helicopter turboshaft engine parameters, which are regulated based on the filtering method. This made achieving almost 100% (0.995 or 99.5%) accuracy possible and reduced the loss function to 0.005 (0.5%) after 280 training epochs. An algorithm has been developed for neural network training based on the errors in backpropagation for closed loops, integrating the helicopter turboshaft engine parameters regulated based on the filtering method. It combines increasing the validation set accuracy and controlling overfitting, considering error dynamics, which preserves the model generalization ability. The adaptive training rate improves adaptation to the data changes and training conditions, improving performance. It has been mathematically proven that the helicopter turboshaft engine parameters regulating neural network closed-loop integration using the filtering method, in comparison with traditional filters (median-recursive, recursive and median), significantly improve efficiency. Moreover, that enables reduction of the errors of the 1st and 2nd types: 2.11 times compared to the median-recursive filter, 2.89 times compared to the recursive filter, and 4.18 times compared to the median filter. The achieved results significantly increase the helicopter turboshaft engine sensor readings accuracy (up to 99.5%) and reliability, ensuring aircraft efficient and safe operations thanks to improved filtering methods and neural network data integration. These advances open up new prospects for the aviation industry, improving operational efficiency and overall helicopter flight safety through advanced data processing technologies.

Distributed video transmission reduction approach for energy saving in WMSNs

SummaryWireless Multimedia Sensor Networks (WMSNs) are composed of a large number of sensor nodes that are distributed in a region to collect and transmit data. Video transmission is one of the most important applications of WMSNs because it can provide critical information about monitored areas. WMSNs face challenges related to energy consumption, bandwidth usage, and network congestion related to huge amounts of data collected by sensors. To tackle this problem, this paper proposes the Distributed Video Transmission Reduction Approach for Energy Saving in WMSN (DiViTRA). The method involves two phases: sensing and transmission phases. DiViTRA achieves frame rate adaptation to reduce the number of captured video frames and save energy during the sensing phase. In the transmission phase, three effective techniques, ORB (Oriented FAST and Rotated BRIEF), Brute‐Force (BF) Matcher, and Grid‐based Motion Statistics (GMS) are applied to decide whether to transmit the current captured frame or remove it and adjust the frame capturing rate of the video sensor accordingly. In the case of frame transmission, the DiViTRA approach compresses the frame using two data reduction approaches: PCA (Principal Component Analysis) and Huffman encoding. Through simulations, DiViTRA demonstrates a 12% reduction in energy consumption, and 71% is a ratio of reduction in sent frames while preserving stream quality. The approach has been validated in scenarios involving critical events, showcasing its efficacy in maintaining data integrity during transmission.

System response curve based first‐order optimization algorithms for cyber‐physical‐social intelligence

AbstractThe continuous enhancement of optimization algorithms and their parameters has spurred the expansion of AI into novel application domains such as image recognition and smart home technology. This paper employs the system response curve (SRC) to the adaptive learning rate optimizer, addressing challenges associated with the establishment of the optimizer control model and parameter adjustments affecting the dynamic performance of the system. These insights offer theoretical support for the optimizer's application in deep learning models. To begin, the adaptive learning rate optimizer is a time‐varying system. Based on the intrinsic relationship between the network optimization and the control system, the time domain expression and approximate transfer function of the adaptive learning rate optimizer are derived, and the system dynamic model is established. Furthermore, based on the system control model of the optimizer, it is proposed to explain the performance impacts of different optimizers and their hyperparameters on the deep learning model through the SRC. Finally, experiments are performed on the MNIST, CIFAR‐10, UTKinect‐Action3D, and Florence3D‐Action datasets to validate the control theory of explaining optimizers through system response curves. The experimental results show that the recognition performance of the Adaptive Moment Estimate (Adam) is better than that of the Adaptive Gradient (AdaGrad) and Root Mean Square Propagation (RMSprop). Additionally, the learning rate affects the model training speed, and the practical application aligns with the theoretical analysis.

Robust Online Estimation of State of Health for Lithium-Ion Batteries Based on Capacities under Dynamical Operation Conditions

Lithium-ion batteries, as the main energy storage component of electric vehicles (EVs), play a crucial role in ensuring the safe and reliable operation of the battery systems through monitoring their state of health (SOH). However, temperature variations and battery aging have significant impacts on the internal parameters of lithium-ion batteries, and these changes directly correlate with the accuracy of battery SOH estimation. To address these issues, this paper proposes an estimation method for lithium-ion battery SOH that considers the impact of temperature. The method begins with reconstructing a second-order hybrid equivalent circuit model for lithium-ion batteries, through which an adaptive update rate for battery model parameters is designed. On this basis, a nonlinear observer for battery states is introduced by integrating filters to estimate SOH. The proposed method considers the impact of capacity in the design of the parameter adaptive update rate, enabling the capacity to be dynamically adjusted based on the actual state of the batteries. This reduces the cumulative error in the SOC observer and improves the modeling accuracy of battery models. Experimental results demonstrate that the method proposed in this paper exhibits exceptional performance in SOH estimation under different temperature conditions. The mean absolute error for SOH estimation does not exceed 0.5%, and the root mean square error does not exceed 0.2%. This method can significantly improve the estimation accuracy of SOH, providing a more efficient and accurate solution for battery management systems in EVs.

Triple-diode solar cell current optimization – An analytical solution based on the Lambert W function

A novel mathematical modeling approach for the Triple-Diode Model (TDM) and its parameter estimation is introduced in this paper. For the first time in the literature, this paper presents two analytical solutions for calculating the current in TDM solar cells using the Lambert W function. Additionally, a new metaheuristic algorithm, the Adaptive Evaporation Rate Water Cycle Algorithm (Adaptive ERWCA), is proposed for estimating TDM solar cell parameters. The proposed algorithm is tested using both the new analytical solutions and a well-known iterative approach for the current-voltage characteristics of the TDM solar cells model. Extensive testing is conducted on three well-known solar cells/modules and a panel under various weather conditions. The results demonstrate that the proposed analytical expressions for the current-voltage characteristics are accurate and highly effective for representing TDM solar cell current. The accuracy of the parameter estimates obtained with the application of the proposed algorithm, considering both analytical solutions and the iterative approach, is compared with existing literature results. It is shown that the proposed algorithm outperforms existing algorithms in solar cell parameter estimation across both the analytical and iterative approaches for current-voltage characteristics. Remarkably, for the RTC France solar cell, the SOLAREX MSX 60 solar module, and the PHOTOWATT-PWP 201 solar module, the obtained Root Mean Square Errors (RMSE) are 0.00075141, 0.01157153, and 0.0010578, respectively. These results are currently the best reported in the literature.

Packet Drop Attack Prevention in DRL Based Data Rate Adaptation Scheme for MANET

Packet Drop Attack Prevention in DRL Based Data Rate Adaptation Scheme for MANET

Improved optimization for the neural-network quantum states and tests on the chromium dimer.

The advent of Neural-network Quantum States (NQS) has significantly advanced wave function ansatz research, sparking a resurgence in orbital space variational Monte Carlo (VMC) exploration. This work introduces three algorithmic enhancements to reduce computational demands of VMC optimization using NQS: an adaptive learning rate algorithm, constrained optimization, and block optimization. We evaluate the refined algorithm on complex multireference bond stretches of H2O and N2 within the cc-pVDZ basis set and calculate the ground-state energy of the strongly correlated chromium dimer (Cr2) in the Ahlrichs SV basis set. Our results achieve superior accuracy compared to coupled cluster theory at a relatively modest CPU cost. This work demonstrates how to enhance optimization efficiency and robustness using these strategies, opening a new path to optimize large-scale restricted Boltzmann machine-based NQS more effectively and marking a substantial advancement in NQS's practical quantum chemistry applications.

Time is NTCP: Should we maximize patient throughput or perform online adaptation on proton therapy systems?

BackgroundCompared to conventional radiotherapy (XT), proton therapy (PT) may improve normal tissue complication probabilities (NTCP). However, PT typically requires higher adaptation rates due to an increased sensitivity to anatomical changes. Systematic online adaptation may address this issue, but it requires additional replanning time, decreasing patient throughput. Therefore, less patients would benefit in such case from PT for a given machine capacity, with results in worse NTCP. AimTo investigate the trade-off between PT patient throughput and NTCP gain as a function of the time needed for adaptation. MethodsA retrospective database of 14 lung patients with two repeated 4DCTs was used to compare NTCP values between XT and PT for NTCP2ym (2-year mortality), NTCPdysphagia and NTCPpneumonitis. Four scenarios were considered for PT: no adaptation using clinical robustness parameters (4D robust optimization, 3 % range error and PTV-equivalent setup errors); systematic online adaptation with clinical robustness parameters; setup errors reduced to 4 mm and to 2 mm. Dose was accumulated on the planning CT. The number of patients treated with PT depended on the extra time needed for adaptation, assuming an 8-hours capacity (assuming 14 patients a day; thus minimum 34.2 min per treatment session if there is no or instantaneous adaptation). ResultsBaseline NTCP gains (PT against XT without adaptation) equaled 6.9 %, 6.1 %, and 7.7 % for NTCP2ym, NTCPdysphagia and NTCPpneumonitis, respectively. Using instantaneous online adaptation and setup errors of 2 mm, the overall gains were then 10.7 %, 13.6 % and 12.4 %. Taking into account loss of capacity, 13.7 min was the maximum extra-time allowed to complete adaptation and maintain an advantage on all three metrics for the 2-mm setup error scenario. ConclusionThis study highlights the critical importance of keeping short online adaptation times when using systems with limited capacity like PT.

Exploring Electrocortical Signatures of Gait Adaptation: Differential Neural Dynamics in Slow and Fast Gait Adapters.

Individuals exhibit significant variability in their ability to adapt locomotor skills, with some adapting quickly and others more slowly. Differences in brain activity likely contribute to this variability, but direct neural evidence is lacking. We investigated individual differences in electrocortical activity that led to faster locomotor adaptation rates. We recorded high-density electroencephalography while young, neurotypical adults adapted their walking on a split-belt treadmill and grouped them based on how quickly they restored their gait symmetry. Results revealed unique spectral signatures within the posterior parietal, bilateral sensorimotor, and right visual cortices that differ between fast and slow adapters. Specifically, fast adapters exhibited lower alpha power in the posterior parietal and right visual cortices during early adaptation, associated with quicker attainment of steady-state step length symmetry. Decreased posterior parietal alpha may reflect enhanced spatial attention, sensory integration, and movement planning to facilitate faster locomotor adaptation. Conversely, slow adapters displayed greater alpha and beta power in the right visual cortex during late adaptation, suggesting potential differences in visuospatial processing. Additionally, fast adapters demonstrated reduced spectral power in the bilateral sensorimotor cortices compared with slow adapters, particularly in the theta band, which may suggest variations in perception of the split-belt perturbation. These findings suggest that alpha and beta oscillations in the posterior parietal and visual cortices and theta oscillations in the sensorimotor cortex are related to the rate of gait adaptation.

Within-Host Viral Growth and Immune Response Rates Predict Foot-and-Mouth Disease Virus Transmission Dynamics for African Buffalo.

AbstractInfectious disease dynamics operate across biological scales: pathogens replicate within hosts but transmit among populations. Functional changes in the pathogen-host interaction thus generate cascading effects across organizational scales. We investigated within-host dynamics and among-host transmission of three strains (SAT-1, -2, -3) of foot-and-mouth disease viruses (FMDVs) in their wildlife host, African buffalo. We combined data on viral dynamics and host immune responses with mathematical models to ask the following questions: How do viral and immune dynamics vary among strains? Which viral and immune parameters determine viral fitness within hosts? And how do within-host dynamics relate to virus transmission? Our data reveal contrasting within-host dynamics among viral strains, with SAT-2 eliciting more rapid and effective immune responses than SAT-1 and SAT-3. Within-host viral fitness was overwhelmingly determined by variation among hosts in immune response activation rates but not by variation among individual hosts in viral growth rate. Our analyses investigating across-scale linkages indicate that viral replication rate in the host correlates with transmission rates among buffalo and that adaptive immune activation rate determines the infectious period. These parameters define the virus's relative basic reproductive number (ℛ0), suggesting that viral invasion potential may be predictable from within-host dynamics.