Memory Architecture Research Articles

BiLSTM-CNN with fixed weight approach for tracking speech articulatory features

Speech production is a complex sequential process which involve the coordination of various articulatory features. Among them tongue being a highly versatile active articulator responsible for shaping airflow to produce targeted speech sounds that are intellectual, clear, and distinct. This paper presents a novel approach for predicting tongue and lip articulatory features involved in a given speech acoustics using a stacked Bidirectional Long Short-Term Memory (BiLSTM) architecture, combined with a one-dimensional Convolutional Neural Network (CNN) for post-processing with fixed weights initialization. The proposed network is trained with two datasets consisting of simultaneously recorded speech and Electromagnetic Articulography (EMA) datasets, each introducing variations in terms of geographical origin, linguistic characteristics, phonetic diversity, and recording equipment. The performance of the model is assessed in Speaker Dependent (SD), Speaker Independent (SI), corpus dependent (CD) and cross corpus (CC) modes. Experimental results indicate that the proposed model with fixed weights approach outperformed the adaptive weights initialization with in relatively minimal number of training epochs. These findings contribute to the development of robust and efficient models for articulatory feature prediction, paving the way for advancements in speech production research and applications.

Optimizing code allocation for hybrid on-chip memory in IoT systems

With the increasing application of IoT devices, the memory subsystem, as the performance and energy bottleneck of IoT systems, has received a lot of attention. One of the keys is on-chip memory which can bridge the performance gap between the CPU and main memory. While many off-the-shelf embedded processors utilize the hybrid on-chip memory architecture containing scratchpad memories (SPMs) and caches, most existing literature ignores the collaboration between caches and SPMs. This paper proposes static SPM allocation strategies for the architecture mentioned above in IoT systems, which try to minimize the overall instruction memory subsystem latency and/or energy consumption. We capture the intra- and inter-task cache conflict misses via a fine-grained temporal cache behavior model. Based on this cache conflict information, we propose an integer linear programming (ILP) algorithm to generate an optimal static function level SPM allocation for system performance. Furthermore, to improve the scalability of the proposed allocation scheme for an enormous task set, we offer the interference factor to calculate the interference impact quantitatively. Then, based on the interference factor, we present two approximate knapsack based heuristic algorithms to provide near optimal static allocation schemes at both function- and basic block-level granularities, which favors fast design space exploration. The experiment results demonstrate that the proposed solution achieves a 30.85% improvement in memory performance, and up to 31.39% reduction in energy consumption, compared to the existing SPM allocation scheme at the function level. In addition, the proposed basic block level allocation algorithm shows better performance than our function level allocation algorithm and other basic block level allocation algorithm.

Brain-Inspired Agents for Quantum Reinforcement Learning

In recent years, advancements in brain science and neuroscience have significantly influenced the field of computer science, particularly in the domain of reinforcement learning (RL). Drawing insights from neurobiology and neuropsychology, researchers have leveraged these findings to develop novel mechanisms for understanding intelligent decision-making processes in the brain. Concurrently, the emergence of quantum computing has opened new frontiers in artificial intelligence, leading to the development of quantum machine learning (QML). This study introduces a novel model that integrates quantum spiking neural networks (QSNN) and quantum long short-term memory (QLSTM) architectures, inspired by the complex workings of the human brain. Specifically designed for reinforcement learning tasks in energy-efficient environments, our approach progresses through two distinct stages mirroring sensory and memory systems. In the initial stage, analogous to the brain’s hypothalamus, low-level information is extracted to emulate sensory data processing patterns. Subsequently, resembling the hippocampus, this information is processed at a higher level, capturing and memorizing correlated patterns. We conducted a comparative analysis of our model against existing quantum models, including quantum neural networks (QNNs), QLSTM, QSNN and their classical counterparts, elucidating its unique contributions. Through empirical results, we demonstrated the effectiveness of utilizing quantum models inspired by the brain, which outperform the classical approaches and other quantum models in optimizing energy use case. Specifically, in terms of average, best and worst total reward, test reward, robustness, and learning curve.

Quantum Random Access Memory Architectures Using 3D Superconducting Cavities

Quantum random access memory (QRAM) is a common architecture resource for algorithms with many proposed applications, including quantum chemistry, windowed quantum arithmetic, unstructured search, machine learning, and quantum cryptography. Here, we propose two bucket-brigade QRAM architectures based on high-coherence superconducting resonators, which differ in their realizations of the conditional-routing operations. In the first, we directly construct cavity-controlled controlled-SWAP (CSWAP) operations, while in the second, we utilize the properties of giant-unidirectional emitters (GUEs). For both architectures, we analyze single- and dual-rail implementations of a bosonic qubit. In the single-rail encoding, we can detect first-order ancilla errors, while the dual-rail encoding additionally allows for the detection of photon losses. For parameter regimes of interest, the postselected infidelity of a QRAM query in a dual-rail architecture is nearly an order of magnitude below that of a corresponding query in a single-rail architecture. These findings suggest that dual-rail encodings are particularly attractive as architectures for QRAM devices in the era before fault tolerance. Published by the American Physical Society 2024

Long-term solar radiation forecasting based on LSTM and attention mechanism: a case study in Algeria

The growing adoption of photovoltaic (PV) energy in urban areas underscores its capability to meet energy demands effectively. The accurate forecasting of meteorological parameters, particularly global solar radiation, is paramount for the efficient management and utilization of solar energy resources. Solar radiation forecasting methods typically fall into two main categories: cloud imagery combined with physical approaches, and Artificial Intelligence (AI) based methods. Due to the non-stationary nature of solar radiation and the high nonlinearity of atmospheric conditions, conventional forecasting approaches often exhibit poor accuracy. Artificial intelligence based approaches, such as machine learning and deep learning algorithms, are extensively employed in global solar radiation forecasting research, demonstrating impressive accuracy. In this respect, this paper presents a novel approach for long term monthly forecasting of global solar radiation using a Bidirectional Long Short-Term Memory (LSTM) architecture augmented with an attention mechanism that includes a Squeeze and Excitation (SE) block (SE-BiLSTM). The effectiveness of this model is extensively evaluated for long-term monthly solar radiation forecasting using meteorological data collected over a 20-year period (from Jan 2001 to Dec 2020) from the National Aeronautics and Space Administration (NASA). The proposed SE-BiLSTM model is compared with well-established forecasting models including Naïve, Autoregressive Moving Average (ARMA), Multi-layer Perceptron (MLP), LSTM, and Bidirectional LSTM. Through rigorous simulation tests, our model demonstrates superior performance, achieving the lowest mean absolute percentage error (MAPE) of 4.52% and mean absolute error (MAE) of 7.89 kW/m2. This advancement holds significant promise for enhancing solar energy forecasting accuracy and its practical application in renewable energy systems.

A memory and attention-based reinforcement learning for musculoskeletal robots with prior knowledge of muscle synergies

PurposeLimited by the types of sensors, the state information available for musculoskeletal robots with highly redundant, nonlinear muscles is often incomplete, which makes the control face a bottleneck problem. The aim of this paper is to design a method to improve the motion performance of musculoskeletal robots in partially observable scenarios, and to leverage the ontology knowledge to enhance the algorithm’s adaptability to musculoskeletal robots that have undergone changes.Design/methodology/approachA memory and attention-based reinforcement learning method is proposed for musculoskeletal robots with prior knowledge of muscle synergies. First, to deal with partially observed states available to musculoskeletal robots, a memory and attention-based network architecture is proposed for inferring more sufficient and intrinsic states. Second, inspired by muscle synergy hypothesis in neuroscience, prior knowledge of a musculoskeletal robot’s muscle synergies is embedded in network structure and reward shaping.FindingsBased on systematic validation, it is found that the proposed method demonstrates superiority over the traditional twin delayed deep deterministic policy gradients (TD3) algorithm. A musculoskeletal robot with highly redundant, nonlinear muscles is adopted to implement goal-directed tasks. In the case of 21-dimensional states, the learning efficiency and accuracy are significantly improved compared with the traditional TD3 algorithm; in the case of 13-dimensional states without velocities and information from the end effector, the traditional TD3 is unable to complete the reaching tasks, while the proposed method breaks through this bottleneck problem.Originality/valueIn this paper, a novel memory and attention-based reinforcement learning method with prior knowledge of muscle synergies is proposed for musculoskeletal robots to deal with partially observable scenarios. Compared with the existing methods, the proposed method effectively improves the performance. Furthermore, this paper promotes the fusion of neuroscience and robotics.

Investigation of Load, Solar and Wind Generation as Target Variables in LSTM Time Series Forecasting, Using Exogenous Weather Variables

Accurately forecasting energy metrics is essential for efficiently managing renewable energy generation. Given the high variability in load and renewable energy power output, this represents a crucial area of research in order to pave the way for increased adoption of low-carbon energy solutions. Whilst the impact of different neural network architectures and algorithmic approaches has been researched extensively, the impact of utilising additional weather variables in forecasts have received far less attention. This article demonstrates that weather variables can have a significant influence on energy forecasting and presents methodologies for using these variables within a long short-term memory (LSTM) architecture to achieve improvements in forecasting accuracy. Moreover, we introduce the use of the seasonal components of the target time series, as exogenous variables, that are also observed to increase accuracy. Load, solar and wind generation time series were forecast one hour ahead using an LSTM architecture. Time series data were collected in five Spanish cities and aggregated for analysis, alongside five exogenous weather variables, also recorded in Spain. A variety of LSTM architectures and hyperparameters were investigated. By tuning exogenous weather variables, a 33% decrease in mean squared error was observed for solar generation forecasting. A 22% decrease in mean absolute squared error (MASE), compared to 24-h ahead forecasts made by the Transmission Service Operator (TSO) in Spain, was also observed for solar generation. Compared to using the target variable in isolation, utilising exogenous weather variables decreased MASE by approximately 10%, 15% and 12% for load, solar and wind generation, respectively. By using the seasonal component of the target variables as an exogenous variable itself, we demonstrated decreases in MASE of 19%, 12% and 8% for load, solar and wind generation, respectively. These results emphasise the significant benefits of incorporating weather and seasonal components into energy-related time series forecasts.

Utilizing BiLSTM For Fine-Grained Aspect-Based Travel Recommendations Using Travel Reviews In Low Resourced Language

Recommender systems have become an essential tool for enhancing user experiences by providing personalized recommendations. In this study, we present a novel approach to constructing a recommender system specifically tailored for Malayalam travel reviews. Our objective was to extract relevant features from these reviews and employ a bidirectional Long Short-Term Memory (BiLSTM) architecture to construct a robust and accurate recommendation model. We focused on four key features extracted from the travel reviews: travel mode, travel type, location climate, and location type. The travel mode feature encompassed the mode of transport opted for the travel such as bus, car, train, etc., while the travel type captured the nature of the trip, including family, friends, or solo travel. Additionally, we considered the climate of the location, including rainy, snowy, hot, and dry, among others, and the location type, such as beach, hilly, or forest destinations. To construct our recommender system, we implemented a BiLSTM architecture, a powerful deep-learning model known for effectively capturing temporal dependencies in sequential data. This architecture allowed us to process the extracted features and learn the underlying patterns within the Malayalam travel reviews. Our experiments were conducted on a comprehensive dataset of Malayalam travel reviews, carefully curated for this study. The dataset encompassed a diverse range of travel experiences, enabling our model to learn from a wide variety of user preferences and recommendations. The performance evaluation of our recommender system yielded promising results. With an accuracy of 83.65 percent, our model showcased its ability to accurately predict and recommend travel options based on the extracted features from the reviews. The high accuracy achieved by our model underscores the effectiveness of the BiLSTM architecture in capturing the nuances of the Malayalam language and understanding the subtle preferences expressed in travel reviews. The practical implications of our work are significant, as it offers a valuable tool for travelers seeking personalized recommendations based on their travel preferences. The use of the Malayalam language in this context expands the reach of recommender systems to a wider audience, catering specifically to individuals who prefer to consume content and make decisions in their native language.

A robust neural network for privacy-preserving heart rate estimation in remote healthcare systems

In this study, we propose a computationally-light and robust neural network for estimating heart rate in remote healthcare systems. We develop a model that can be trained on consumer-grade graphics processing units (GPUs), and can be deployed on edge devices for swift inference. We propose a hybrid model based on convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) architectures for estimating heart rate from Electrocardiogram (ECG) and Photoplethysmography (PPG) signals. Considering the sensitive nature of the ECG signals, we ensure a formal privacy guarantee, differential privacy, for the model training. We perform a tight accounting of the overall privacy budget of our training algorithm using the Rényi Differential Privacy technique. We demonstrate that our model outperforms state-of-the-art networks on a benchmark dataset for both ECG and PPG signals despite having a much smaller number of trainable parameters and, consequently, much smaller training and inference times. Our CNN-BiLSTM architecture can also provide excellent heart rate estimation performance even under strict privacy constraints. We develop a prototype Arduino-based data collection system that is low-cost, efficient, and useful for providing access to modern healthcare services to people living in remote areas.

Machine learning and IoT based Predictive Maintenance for the industrial motors for sustained Automation in the power plant Industry

The purpose of this research is to develop novel framework based on advanced tools, including machine learning and the Internet of Things and self-attention mechanisms. Traditional and advanced tools were used in the data-driven predictive maintenance mechanism and served as a basis for comparing the tools. At the power plant, thirty-four datasets were collected to monitor three industrial motors continuously. The tools’ predictive ability was analysed using conceptualized features from the sensory data , and the management strategy remained dependent on the network. The study results show that each tool performs at high-performing levels because each exceeded 75% of the performance metric. The study results indicated that the proposed framework gave a consistent high-performance metric, 86.4%, in all the ten experimental scenarios used . The rate determination was attributed to the choice of the long short-term memory architecture, the self-attention mechanisms, and the optimization techniques. The choice of using mean squared logarithmic error also contributed to the outcomes because these tools yield high-performance scores. In addition to these, the test of the forecasting models chosen also influenced this performance. This study’s findings show that this framework is reliable in predicting pending failure equipment and developing relevant management strategies in industrial motors failure. It was found that the use of advanced tools in the development of the framework was a crucial aspect. Scaled tools and optimization techniques are of crucial importance in predictive management in equipment failure as they assist the frameworks in identifying ideal features from the sensory data. It should also be noted that the choice of a loss-function model is also important in the predictability of framework..

Deepfake Detection Based on Temporal Analysis of Facial Dynamics Using LSTM and ResNeXt Architectures

The proliferation of deepfake technology presents a critical challenge to the authenticity and trustworthiness of digital media. To address this issue, we propose an innovative deepfake detection framework that combines the power of Long Short-Term Memory (LSTM) and ResNeXt architectures. By integrating spatial and temporal analysis methods, our approach aims to accurately identify manipulated videos amidst the vast sea of online content. Through rigorous experimentation and evaluation using diverse datasets, our framework demonstrates promising results in effectively distinguishing between genuine and fake videos. This research contributes to the ongoing efforts to combat deepfake misinformation and uphold the integrity of digital media platforms.

The Significance of Dynamic COVID-19 Dashboard in Formulating School Reopening Strategies

Experiments conducted with the COVID-19 dataset have predominantly concentrated on predicting cases fluctuating and classifying lung-related diseases. Nevertheless, the consequences of the COVID-19 pandemic have also spread to the education sector. To safeguard educational stability in response to the remote learning policy, we leverage authentic COVID-19 datasets alongside school information across 154 sub-areas in Surabaya City, Indonesia. Our focus is predicting the dynamic within these sub-areas where schools are located. The outcomes of this study, by incorporating the recurrent neural network of long- and short-term memory (RNN-LSTM) architecture and refined hyperparameters, effectively enhanced the predictive model's performance. The findings are showcased on a dashboard, visually representing the transmission of COVID-19 in schools across each sub-area. This information serves as a basis for informed decisions on the safe reopening of schools, aiming to mitigate the decline in education quality during the challenging pandemic.

ARCHITECTURAL AND STRUCTURAL AND FUNCTIONAL FEATURES OF THE ORGANIZATION OF PARALLEL-HIERARCHICAL MEMORY

Parallel hierarchical memory (PI memory) is a new type of memory that is designed to improve the performance of parallel computing systems. PI memory is composed of two blocks: a mask RAM and a tail element RAM. The mask RAM stores the masks that are used to encode the information, while the tail element RAM stores the actual information. The address block of the PI memory is responsible for generating the physical addresses of the cells where the tail elements and their masks are stored. The address block also stores the field of addresses where the array was written and associates this field of addresses with the corresponding external address used to write the array. The proposed address block structure is able to efficiently generate the physical addresses of the cells where the tail elements and their masks are stored. The address block is also able to store the field of addresses where the array was written and associate this field of addresses with the corresponding external address used to write the array. The proposed address block structure has been implemented in a prototype PI memory. The prototype PI memory has been shown to be able to achieve significant performance improvements over traditional memory architectures. The paper will present a detailed description of the PI transformation algorithm, a description of the different modes of addressing organization that can be used in PI memory, an analysis of the efficiency of parallel-hierarchical memory structures, and a discussion of the challenges and future research directions in the field of PI memory.

Enhancing Spam Detection: A Crow-Optimized FFNN with LSTM for Email Security

Email security is paramount in today's digital landscape, as the proliferation of spam emails poses a significant threat to individuals and organizations alike. To combat this menace, this study introduces a novel approach that marries the power of Crow Search Optimization (CSO) with a Feedforward Neural Network (FFNN) and Long Short-Term Memory (LSTM) architecture to bolster spam detection. The proposed Crow-Optimized FFNN with LSTM (C-FFNN-LSTM) leverages CSO to fine-tune the neural network's parameters, optimizing its ability to distinguish between legitimate emails and spam. The CSO algorithm mimics the collaborative behavior of crows, thereby enhancing the model's convergence and robustness. Experimental results showcase the effectiveness of the C-FFNN-LSTM approach, achieving remarkable accuracy rates and reducing false positives. This innovation not only enhances email security but also offers a promising avenue for refining spam detection algorithms across various domains. In an era of ever-evolving cyber threats, the C-FFNN-LSTM framework stands as a beacon of improved email security, safeguarding digital communication channels.
 In our methodology, we attained an outstanding accuracy level of 99.1% during the testing phase.

Short-term wind power forecasting through stacked and bi directional LSTM techniques.

Computational intelligence (CI) based prediction models increase the efficient and effective utilization of resources for wind prediction. However, the traditional recurrent neural networks (RNN) are difficult to train on data having long-term temporal dependencies, thus susceptible to an inherent problem of vanishing gradient. This work proposed a method based on an advanced version of RNN known as long short-term memory (LSTM) architecture, which updates recurrent weights to overcome the vanishing gradient problem. This, in turn, improves training performance. The RNN model is developed based on stack LSTM and bidirectional LSTM. The parameters like mean absolute error (MAE), standard deviation error (SDE), and root mean squared error (RMSE) are utilized as performance measures for comparison with recent state-of-the-art techniques. Results showed that the proposed technique outperformed the existing techniques in terms of RMSE and MAE against all the used wind farm datasets. Whereas, a reduction in SDE is observed for larger wind farm datasets. The proposed RNN approach performed better than the existing models despite fewer parameters. In addition, the approach requires minimum processing power to achieve compatible results.

Arabic text summarization using three-layer bidire ctional long short-term memory (BiLSTM) architecture

This work presents an improved approach to the challenging problem of Arabic Text Summarization (ATS) by introducing a novel model that seamlessly integrates state-of-the-art neural network architectures with advanced Natural Language Processing (NLP) techniques. Inspired by classical ATS approaches, our model leverages a three-layer Bidirectional Long Short-Term Memory (BiLSTM) architecture which is augmented with Transformer-based attention mechanisms and AraBERT for preprocessing, to successfully tackle the notoriously challenging peculiarities of the Arabic language. To boost performance, our model further draws upon the power of contextual embeddings from models such as GPT-3, and through the use of advanced data augmentation techniques including back-translation and paraphrasing. To further improve performance, our approach integrates novel techniques for training and uses Bayesian Optimization to perform hyperparameter optimization. The model evaluated against state-of-the-art datasets such as the Arabic Headline Summary (AHS) and Arabic Mogalad_Ndeef (AMN) and reported on traditional evaluation metrics including: ROUGE-1, ROUGE-2, ROUGE-L, BLEU, METEOR and BERTScore. This work is significant because it presents an important step forward in the task of Arabic Text Summarization (ATS) towards summarizing text to be not only coherent and concise, but also authentic and culturally relevant in an effort to push forward NLP research and applications for Arabic. Представлен улучшенный подход к сложной проблеме арабской суммаризации текста (Arabic Text Summarization – ATS) путем внедрения новой модели, которая бесшовно интегрирует передовые архитектуры нейронных сетей с продвинутыми техниками обработки естественного языка (Natural Language Processing – NLP). Вдохновляясь классическими подходами ATS, представленная модель использует трехуровневую архитектуру двунаправленной сети с длинной краткосрочной памятью (Bidirectional Long Short-Term Memory – BiLSTM), которая дополнена механизмами внимания на основе трансформера и AraBERT для предварительной обработки и успешного решения особенно сложных задач, обусловленных спецификой арабского языка. Для повышения производительности модели также используется мощь контекстуальных вложений от моделей, таких как GPT-3, а также продвинутые техники увеличения данных, включая обратный перевод и перефразирование. Для дальнейшего улучшения производительности при данном подходе интегрируются новые техники обучения и используется байесовская оптимизация для гиперпараметров. Модель оценивалась на передовых наборах данных, таких как арабский сводный заголовок (Arabic Headline Summary – AHS) и арабский могалад_ндеф (Arabic Mogalad_Ndeef – AMN), и отчитывалась по традиционным метрикам оценки, включая ROUGE-1, ROUGE-2, ROUGE-L, BLEU, METEOR и BERTScore. Работа представляет собой важный шаг в решении задачи арабской суммаризации текста (ATS), чтобы он был не только связным и сжатым, но также аутентичным и культурно релевантным в усилиях по продвижению исследований и приложений NLP для арабского языка.

Implementation of Neural Machine Translation in Translating from Indonesian to Sasak Language

Language translation is part of Natural Language Processing, also known as Machine Translation, which helps the process of learning foreign and regional languages using translation technology in sentence form. In Lombok, there are still people who are not very fluent in Indonesian because Indonesian is generally only used at formal events. This research aimed to develop a translation model from Indonesian to Sasak. The method used was the Neural Machine Translation with the Recurrent Neural Network - Long Short Term Memory architecture and the Word2Vec Embedding with a sentence translation system. The dataset used was a parallel corpus from the Tatoeba Project and other open sources, divided into 80% training and 20% validation data. The result of this research was the application of Neural Machine Translation with the Recurrent Neural Network - Long Short Term Memory algorithm, which could produce a model with an accuracy of 99.6% in training data and 71.9% in test data. The highest ROUGE evaluation metric result obtained on the model was 88%. This research contributed to providing a translation model from Indonesian to Sasak for the local community to facilitate communication and preserve regional language culture.

Causal functional maps of brain rhythms in working memory

Human working memory is a key cognitive process that engages multiple functional anatomical nodes across the brain. Despite a plethora of correlative neuroimaging evidence regarding the working memory architecture, our understanding of critical hubs causally controlling overall performance is incomplete. Causal interpretation requires cognitive testing following safe, temporal, and controllable neuromodulation of specific functional anatomical nodes. Such experiments became available in healthy humans with the advance of transcranial alternating current stimulation (tACS). Here, we synthesize findings of 28 placebo-controlled studies (in total, 1,057 participants) that applied frequency-specific noninvasive stimulation of neural oscillations and examined working memory performance in neurotypical adults. We use a computational meta-modeling method to simulate each intervention in realistic virtual brains and test reported behavioral outcomes against the stimulation-induced electric fields in different brain nodes. Our results show that stimulating anterior frontal and medial temporal theta oscillations and occipitoparietal gamma rhythms leads to significant dose-dependent improvement in working memory task performance. Conversely, prefrontal gamma modulation is detrimental to performance. Moreover, we found distinct spatial expression of theta subbands, where working memory changes followed orbitofrontal high-theta modulation and medial temporal low-theta modulation. Finally, all these results are driven by changes in working memory accuracy rather than processing time measures. These findings provide a fresh view of the working memory mechanisms, complementary to neuroimaging research, and propose hypothesis-driven targets for the clinical treatment of working memory deficits.

DFDTA-MultiAtt: Multi-attention based deep learning ensemble fusion network for drug target affinity prediction

<p>An essential step in the drug development process is the accurate detection of drug-target interactions (DTI). The importance of binding affinity values in understanding protein-ligand interactions was previously disregarded, and DTI prediction was only seen as a binary classification problem. In this regard, we introduced the DFDTA-MultiAtt model for predicting the drug target binding affinity in two stages using the structural and sequential information. The first step of the first stage involves retrieving features from sequence data using a bi-directional long short term memory (Bi-LSTM) architecture together with a multi-attention module and dilated convolutional neural network (dilated-CNN) architecture, and the second step features are learnt from structure representation once again using a dilated-CNN. To predict the binding affinity, the second stage uses an ensemble learning model. The proposed model also produces findings with a greater overall accuracy when compared to contemporary state-of-the-art methods. The model generates an enormous +0.006 concordance index (CI) score on the Davis dataset and reduces the mean square error (MSE) by 0.174 on the KIBA dataset.</p>

A Proactive Self-Adaptation Approach Based on Ensemble Prediction for Service-Based Systems

Service-based systems (SBSs) are a unique category of software systems that dynamically combine various third-party services at runtime to deliver complex and adaptive functionality. This dynamic composition introduces a high level of unpredictability and uncertainty, creating potential anomalies and exerting significant pressure on system maintenance. To tackle this challenge, the conventional approach involves employing prediction-based proactive self-adaptation. Despite the prevalence of existing approaches emphasizing prediction accuracy, the critical aspect of “earliness” in predictions is often overlooked. Striking a balance between early and accurate predictions is paramount in practice. In response, we propose Proactive Self-Adaptation based on Ensemble Prediction (PSA-EP) to effectively balance the trade-off between the prediction earliness and accuracy of in SBSs. At the heart of PSA-EP lies an ensemble prediction model built upon a deep neural network and an enhanced long short-term memory (DNN-ELSTM) architecture. PSA-EP is crafted to empower SBSs to adapt to the inherent unpredictability and instability, facilitating the achievement of their adaptation goals. This adaptation mechanism not only enables effective prediction and analysis of adaptation goal violations but also addresses the reliability and performance of service level agreements (SLAs) governing quality of service (QoS). We evaluated the performance of PSA-EP in a decentralized tele-assistance system using four key metrics, and the experimental results, examined from various perspectives, underscore its exceptional performance.