DF-BETA: An FPGA-based Memory Locality Aware Decision Forest Accelerator via Bit-Level Early Termination

This study aimed to construct a high-performance prediction and diagnosis model for type 2 diabetic retinopathy (DR) and identify key correlates of DR. This study utilized a cross-sectional dataset of 3,000 patients from the People’s Liberation Army General Hospital in 2021. Logistic regression was used as the baseline model to compare the prediction performance of the machine learning model and the related factors. The recursive feature elimination cross-validation (RFECV) algorithm was used to select features. Four machine learning models, support vector machine (SVM), decision tree (DT), random forest (RF), and gradient boost decision tree (GBDT), were developed to predict DR. The models were optimized using grid search to determine hyperparameters, and the model with superior performance was selected. Shapley-additive explanations (SHAP) were used to analyze the important correlation factors of DR. Among the four machine learning models, the optimal model was GBDT, with predicted accuracy, precision, recall, F1-measure, and AUC values of 0.7883, 0.8299, 0.7539, 0.7901, and 0.8672, respectively. Six key correlates of DR were identified, including rapid micronutrient protein/creatinine measurement, 24-h micronutrient protein, fasting C-peptide, glycosylated hemoglobin, blood urea, and creatinine. The logistic model had 27 risk factors, with an AUC value of 0.8341. A superior prediction model was constructed that identified easily explainable key factors. The number of correlation factors was significantly lower compared to traditional statistical methods, leading to a more accurate prediction performance than the latter.

Linear rankSVM is one of the widely used methods for learning to rank. Although its performance may be inferior to nonlinear methods such as kernel rankSVM and gradient boosting decision trees, linear rankSVM is useful to quickly produce a baseline model. Furthermore, following its recent development for classification, linear rankSVM may give competitive performance for large and sparse data. A great deal of works have studied linear rankSVM. The focus is on the computational efficiency when the number of preference pairs is large. In this letter, we systematically study existing works, discuss their advantages and disadvantages, and propose an efficient algorithm. We discuss different implementation issues and extensions with detailed experiments. Finally, we develop a robust linear rankSVM tool for public use.

The accurate estimation of caloric density in food products is a critical component of nutritional science and dietary management, yet experimental determination remains resource intensive. The research develops a robust computational framework for predicting caloric energy based on standard nutritional composition variables using advanced machine learning techniques. To achieve this, a dataset comprising 410 food items with seven predictors, including protein, fat, carbohydrates, sugar, dietary fiber, sodium, and potassium, was utilized to train a Gradient Boosting Decision Trees (GBDT) model. The study evaluated the efficacy of four exceptional hyperparameter optimization algorithms: BBO, BPI, GPO, and evolutionary strategies (ES). Performance was rigorously assessed using 5‐fold cross‐validation and statistical metrics including R 2 , MSE, and AARE%. The results demonstrated that the GBDT‐ES configuration achieved the best performance with a test R 2 of 0.982950 and an AARE% of 3.661596%, whereas GBDT‐GPO offered a competitive balance of accuracy and computational efficiency with the lowest runtime. Furthermore, SHAP analysis revealed that carbohydrates and fats were the primary drivers of caloric estimation, ensuring the model aligned with biological energy densities. In conclusion, the integration of evolutionary optimization with gradient boosting provides a highly precise and scientifically interpretable tool for nutritional analysis, offering a viable alternative to traditional laboratory calorimetry.

Water is a vital resource, and its quality has a direct impact on human health. Groundwater, as one of the primary water sources, requires careful monitoring to ensure its safety. Although manual methods for testing water quality are accurate, they are often time-consuming, costly, and inefficient when dealing with large and complex data sets. In recent years, machine learning has become an effective alternative for water quality assessment. However, current approaches still face challenges, such as the limited performance of individual models, minimal improvements from optimization algorithms, lack of dynamic feature weighting mechanisms, and potential information loss when simplifying model inputs. To address these challenges, this paper proposes a hybrid model, BS-MLP, which combines GBDT (gradient-boosted decision tree) and MLP (multilayer perceptron). The model leverages GBDT’s strength in feature selection and MLP’s capability to manage nonlinear relationships, enabling it to capture complex interactions between water quality parameters. We employ Bayesian optimization to fine-tune the model’s parameters and introduce a feature-weighting attention mechanism to develop the BS-FAMLP model, which dynamically adjusts feature weights, enhancing generalization and classification accuracy. In addition, a comprehensive parameter selection strategy is employed to maintain data integrity. These innovations significantly improve the model’s classification performance and efficiency in handling complex water quality environments and imbalanced datasets. This model was evaluated using a publicly available groundwater quality dataset consisting of 188,623 samples, each with 15 water quality parameters and corresponding labels. The BS-FAMLP model shows strong classification performance, with optimized hyperparameters and an adjusted feature-weighting attention mechanism. Specifically, it achieved an accuracy of 0.9616, precision of 0.9524, recall of 0.9655, F1 Score of 0.9589, and an AUC score of 0.9834 on the test set. Compared to single models, classification accuracy improved by approximately 10%, and when compared to other hybrid models with additional attention mechanisms, BS-FAMLP achieved an optimal balance between classification performance and computational efficiency. The core objective of this study is to utilize the acquired water quality parameter data for efficient classification and assessment of water samples, with the aim of streamlining traditional laboratory-based water quality analysis processes. By developing a reliable water quality classification model, this research provides robust technical support for water safety management.

Prediction of aptamer–protein interacting pairs based on sparse autoencoder feature extraction and an ensemble classifier

Accurate grid frequency prediction is essential for maintaining the stability and reliability of power systems. However, the complex dynamic characteristics of grid frequency and the nonlinear correlations among massive time series data make it challenging for traditional time series prediction methods to balance efficiency and accuracy. In this paper, we propose a Dynamic Significance–Correlation Weighting (D-SCW) method, which generates dynamic weight coefficients that evolve over time. This is achieved by constructing a joint screening mechanism of feature time series correlation analysis and statistical significance test, combined with the LightGBM gradient-boosting decision tree (GBDT) framework; accordingly, high-precision prediction of grid frequency time series data is realized. To verify the effectiveness of the D-SCW method, this study conducts comparative experiments on two actual grid operation datasets (including typical scenarios with wind/photovoltaic (PV) installations, accounting for 5–35% of the grid); additionally, the Spearman’s rank correlation coefficient method, mutual information (MI), Lasso regression, and the feature screening method of recursive feature elimination (RFE) are selected as the baseline control; root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are adopted as assessment indicators. The results show that the D-SCW-LightGBM framework reduces the root mean squared error (RMSE) by 5.2% to 10.4% and shortens the dynamic response delay by 52% compared with the benchmark method in high renewable penetration scenarios, confirming its effectiveness in both prediction accuracy and computational efficiency.

When researchers are faced with building machine learning (ML) radiomic models, the first choice they have to make is what model to use. Naturally, the goal is to use the model with the best performance. But what is the best model? It is well known in ML that modern techniques such as gradient boosting and deep learning have better capacity than traditional models to solve complex problems in high dimensions. Despite this, most radiomics researchers still do not focus on these models in their research. As access to high-quality and large data sets increase, these high-capacity ML models may become even more relevant. In this article, we use a large dataset of 949 prostate cancer patients to compare the performance of a few of the most promising ML models for tabular data: gradient-boosted decision trees (GBDTs), multilayer perceptions, convolutional neural networks, and transformers. To this end, we predict nine different prostate cancer pathology outcomes of clinical interest. Our goal is to give a rough overview of how these models compare against one another in a typical radiomics setting. We also investigate if multitask learning improves the performance of these models when multiple targets are available. Our results suggest that GBDTs perform well across all targets, and that multitask learning does not provide a consistent improvement.

This paper investigates the connections between two state of the art classifiers: decision forests (DFs, including decision jungles) and convolutional neural networks (CNNs). Decision forests are computationally efficient thanks to their conditional computation property (computation is confined to only a small region of the tree, the nodes along a single branch). CNNs achieve state of the art accuracy, thanks to their representation learning capabilities. We present a systematic analysis of how to fuse conditional computation with representation learning and achieve a continuum of hybrid models with different ratios of accuracy vs. efficiency. We call this new family of hybrid models conditional networks. Conditional networks can be thought of as: i) decision trees augmented with data transformation operators, or ii) CNNs, with block-diagonal sparse weight matrices, and explicit data routing functions. Experimental validation is performed on the common task of image classification on both the CIFAR and Imagenet datasets. Compared to state of the art CNNs, our hybrid models yield the same accuracy with a fraction of the compute cost and much smaller number of parameters.

An MRI radiomics approach to predict the efficacy of chemotherapy for osteosarcoma.

Bit-serial computation has been a prevailing convolution method to accelerate varying-precision DNNs by slicing a multi-bit data into multiple 1-bit data and transforming a multiplication into multiple additions, where additions of zero bits are ineffectual, while additions of non-zero bits are repetitive since multiple kernels are quite possible to possess non-zero bits at the same kernel positions. Previous bit-serial accelerators only remove ineffectual additions by skipping computation of zero bits, however, repetitive additions are unable to be eliminated since they compute convolution of each kernel independently. In this work, we propose fused kernel convolution algorithm to eliminate both ineffectual and repetitive additions in bit-serial computation by exploiting bit repetition and bit sparsity in weights, for both convolutional and fully-connected layers. It unifies convolutions of multiple kernels into convolution of one fused kernel by firstly grouping additions into different patterns and secondly reconstructing convolution results, minimizing addition count. Meantime, the memory accesses of activations and partial sums are decreased due to less convolution count. Then a fused kernel convolution based accelerator, FuseKNA, is designed with compact compute logic, which fully exploits value sparsity of activations and bit sparsity of weights. Benchmarked with a set of mainstream DNNs, FuseKNA improves performance by $4.47 \times$, $2.31 \times$ and $1.81 \times$, energy efficiency by $4.13 \times$, $3.06 \times$ and $2.53 \times$ over state-of-the-art Stripes, Pragmatic and Bit-Tactical.

Synthetic Aperture Sonar (SAS) achieves high-resolution imaging of underwater targets by moving a small-aperture array and coherently processing the signals, thereby generating a virtual large-aperture imaging effect. The conventional Range-Doppler (RD) algorithm offers advantages such as high computational efficiency, simple processing flow, and ease of engineering implementation. However, due to the limitations of its underlying assumptions and its heavy reliance on interpolation for range walk correction, the RD algorithm cannot achieve both imaging efficiency and accuracy. To address this issue, this paper proposes an improved RD algorithm based on a sparse Bayesian learning two-dimensional deconvolution (SBL-dCv2) framework. The mathematical model of the algorithm is derived, and comparative simulation experiments with the traditional RD algorithm are conducted for multi-point targets. The simulation results show that the SBL-dCv2 improved RD algorithm delivers both high resolution and computational efficiency, offers stronger resolution performance for multi-point targets, and is better suited for practical engineering applications of SAS.

Event-driven sorting algorithm-based Monte Carlo method with neighbour merging method for solving aerosol dynamics

This paper proposes a new metaheuristic algorithm for scheduling BoT applications, consisting of many independent tasks that can be processed in parallel, in a mobile edge computing environment under mobile device’s energy constraint. We formulate this energy-constrained BoT scheduling problem as an integer program. To solve the formulated optimization problem, we propose an efficient scheduling metaheuristic that employs the whale optimization algorithm (WOA) framework to determine the most appropriate task offloading solution. We use a task sequence to represent the offloading solution. The proposed algorithm starts with using the small-size-first policy to generate the initial solution, which is identified as the best individual in the initial whale population. We develop a cost-effective mapping operator to map each whale individual onto a feasible task sequence. We further design a greedy search-based task assignment strategy to determine the assignment of BoT tasks onto MEC servers under the energy constraint, such that the completion time of BoT applications can be determined and used as the fitness value of the corresponding whale individual. Built upon the WOA framework and DM operator, we develop the metaheuristic algorithm to iteratively update the population and explore better solutions. Experimental results show that the proposed scheduling algorithm outperforms the baseline algorithm in terms of both computational efficiency and scheduling performance when scheduling BoT applications in a MEC environment.

Wireless Sensor Network (WSN) presents significant challenges for the application of distributed signal processing and distributed control. These systems will challenge us to apply appropriate techniques to construct capable processing units with sensing nodes considering energy constraints. Digital Signal Processing (DSP) is one of the capable processing units, but it is not commonly used in WSN because of the power constraint. The Multiply-Accumulate Unit (MAC) is the main computational kernel in DSP architectures. The MAC unit determines the power and the speed of the overall system; it always lies in the critical path. Developing high speed and low power MAC is crucial to use DSP in the future WSN. In this work, a fast and low power MAC Unit is proposed. The proposed architecture is based on examination of the critical delays and hardware complexities of merged MAC architectures to design a unit with a low critical path delay and low hardware complexity. The new architecture reduces the hardware complexity of the summation network, thus reduces the overall power. Increasing the speed of operation is achieved by feeding the bits of the accumulated operand into the summation tree before the final adder instead of going through the entire summation network. The ASIC implementation of the proposed 32-bit MAC unit saves 5.5% of the area, 9% of the energy, and reduces the delay by 13% compared to the regular merged MAC unit.

A computationally efficient approach to obtain vanishing points in images for camera orientation applications is presented in this paper. Modifications are made to the conventional J-linkage algorithm, named Fast J-linkage, so that the detection of vanishing points is achieved in a computationally efficient manner. This fast algorithm is then utilized to recover camera orientation or rotation matrices from video sequences. Both the computational efficiency and performance aspect of the Fast J-linkage algorithm are compared with the conventional J-linkage algorithm by examining the York Urban image database. In addition, the results of an actual camera rotation implementation is provided and validated by using inertial sensors data. The results obtained indicate the higher computational efficiency and accuracy aspects of the developed vanishing point detection.