Weights Of Convolutional Neural Network Research Articles

Plant Disease Detection Using CNN with The Optimization Called Beluga Whale Optimization Mechanism

Plant disease detection is critical for ensuring agricultural productivity. Early and accurate identification of plant diseases can help in the timely application of remedies, reducing yield loss and improving crop quality. This paper presents a deep learning(DL) approach using Convolutional Neural Networks (CNN) for plant disease detection, combined with an advanced optimization technique named the Beluga Whale Optimization Mechanism (BWOM). The CNN is implemented to extract and features from plant images, providing a robust model capable of differentiating between healthy and diseased plants. The BWOM is utilized to optimize the CNN's hyper parameters and weights, enhancing model accuracy and efficiency by reducing overfitting and improving generalization. The BWOM mimics the social behaviour and echolocation techniques of beluga whales to navigate and optimize solution spaces effectively. By iterating through population-based exploration and exploitation phases, BWOM provides a balanced search mechanism to fine-tune CNN parameters. Further the results demonstrate the effectiveness of combining CNN with BWOM in achieving high accuracy rates for plant disease classification.

A novel ensemble learning framework based on a genetic algorithm for the classification of pneumonia

Pneumonia is a disease that can be detected by the opacity changes in chest X-rays and can lead to fatal consequences. Medical image analysis has several challenges, such as limited labeled datasets, imbalanced class distribution, image noise, and overfitting, so individual Convolutional Neural Networks (CNNs) are insufficient to detect pneumonia accurately. Although ensemble CNN models have been used in previous studies, the literature lacks guidance on identifying the optimal CNN models and weight ratio to combine them. In this study, we propose a novel ensemble CNN framework to accurately detect pneumonia, with optimum weights set by a Genetic Algorithm (GA). Firstly, a noise outside the lung was removed, and the model performance was enhanced by performing lung segmentation on Chest X-ray. The performances of several CNN models were analyzed by hyperparameter optimization. The framework combines the three models that give the best accuracy and the two models that provide the lowest false-negative value with the ensemble method in the ratio of the most appropriate weights. The proposed framework provided the best performance on the public test dataset with an accuracy of 97.23% and an F1-score of 97.45% compared to state-of-the-art methods. The study's main contributions are determining suitable models and their optimal weights of the ensemble method based on the GA. The proposed framework enables a rapid and effective diagnostic process, less costly healthcare services, and more efficient use of resources. The demo-link is https://www.youtube.com/watch?v=KZ50K3HL70U.

Epileptic Seizure Detection in EEG Signal Using Optimized Convolutional Neural Network with Selected Feature Set

This article implements novel epileptic seizure detection by EEG signal that includes various phases: Feature extraction, Feature selection, and Classification. Originally, the input EEG signal is subjected to the preprocessing phase, and from the preprocessed signal, certain features are extracted in which the features like TDFs, FDFs, and TFDFs are extracted. Here, the Time Domain Features (TDFs) include Non-linear energy (NE), Permutation Entropy (PE), and Weighted Permutation Entropy (WPE); the Frequency Domain Features (FDFs) consists of Intensity Weighted Mean Frequency (IWMF); and the Time-Frequency Domain Features (TFDFs) consists of the “Weighted Multi-scale Rényi Permutation Entropy (WMRPE)”. Moreover, the extracted features are subjected to the feature selection phase. An improved chi-square model is proposed for selecting the appropriate features. Then, the chosen features are subjected to the classification process. Here, the classification is performed using an Optimized Convolutional Neural Network (CNN). To make the detection more precise, the CNN weights are tuned optimally through a Weighting Factor Based Shark Smell Optimization (IWFSSO) model. The MCC of the implemented CNN+IWFSSO method achieved a better value for a learning percentage 90%; nevertheless, the compared existing models accomplish a smaller value for dataset 2.

Crow Way: An Optimization Technique for generating the Weight and Bias in Deep CNN

Fine-tuning Convolutional Neural Networks (CNNs) weights and biases are essential for solving difficult machine learning problems. We provide a hybrid optimizations strategy that combines the benefits of Crowd Search (CSA) and Brainstorm Optimization (BSO). CSA-BSO optimizes CNNs. CSA-BSO optimizes CNN weights. Our fitness measures determine each crow's location and speed throughout the CSA phase. The BSO phase mixes optimal replies with random ones to create new solutions. The best new ideas move on. After a certain number of iterations, we alternate between the CSA and BSO phases and pick the best solution as the CNN's final state. We design and test the CSA-BSO approach on many CNN architectures and datasets. Experimentally, the hybrid algorithm outperforms CSA and BSO in convergence time and solution quality. CSA and BSO collaborate to efficiently employ feasible regions and investigate a bigger search field, boosting optimizations. Stable and adaptable, CSA-BSO supports many CNN architectures and data formats. This study's main contribution is the CSA-BSO, a unique hybrid optimizations approach for convolutional neural network weight and bias optimizations. It beats CSA and BSO. The recommended strategy improves CNN picture classification, object recognition, and NLP.

Adapting model-based deep learning to multiple acquisition conditions: Ada-MoDL.

The aim of this work is to introduce a single model-based deep network that can provide high-quality reconstructions from undersampled parallel MRI data acquired with multiple sequences, acquisition settings, and field strengths. A single unrolled architecture, which offers good reconstructions for multiple acquisition settings, is introduced. The proposed scheme adapts the model to each setting by scaling the convolutional neural network (CNN) features and the regularization parameter with appropriate weights. The scaling weights and regularization parameter are derived using a multilayer perceptron model from conditional vectors, which represents the specific acquisition setting. The perceptron parameters and the CNN weights are jointly trained using data from multiple acquisition settings, including differences in field strengths, acceleration, and contrasts. The conditional network is validated using datasets acquired with different acquisition settings. The comparison of the adaptive framework, which trains a single model using the data from all the settings, shows that it can offer consistently improved performance for each acquisition condition. The comparison of the proposed scheme with networks that are trained independently for each acquisition setting shows that it requires less training data per acquisition setting to offer good performance. The Ada-MoDL framework enables the use of a single model-based unrolled network for multiple acquisition settings. In addition to eliminating the need to train and store multiple networks for different acquisition settings, this approach reduces the training data needed for each acquisition setting.

Secure and efficient implementation of facial emotion detection for smart patient monitoring system

BackgroundMachine learning has enabled the automatic detection of facial expressions, which is particularly beneficial in smart monitoring and understanding the mental state of medical and psychological patients. Most algorithms that attain high emotion classification accuracy require extensive computational resources, which either require bulky and inefficient devices or require the sensor data to be processed on cloud servers. However, there is always the risk of privacy invasion, data misuse, and data manipulation when the raw images are transferred to cloud servers for processing facical emotion recognition (FER) data. One possible solution to this problem is to minimize the movement of such private data.MethodsIn this research, we propose an efficient implementation of a convolutional neural network (CNN) based algorithm for on‐device FER on a low‐power field programmable gate array (FPGA) platform. This is done by encoding the CNN weights to approximated signed digits, which reduces the number of partial sums to be computed for multiply‐accumulate (MAC) operations. This is advantageous for portable devices that lack full‐fledged resource‐intensive multipliers.ResultsWe applied our approximation method on MobileNet‐v2 and ResNet18 models, which were pretrained with the FER2013 dataset. Our implementations and simulations reduce the FPGA resource requirement by at least 22% compared to models with integer weight, with negligible loss in classification accuracy.ConclusionsThe outcome of this research will help in the development of secure and low‐power systems for FER and other biomedical applications. The approximation methods used in this research can also be extended to other image‐based biomedical research fields.

Deep pre-trained FWI: where supervised learning meets the physics-informed neural networks

SUMMARYFull-waveform inversion (FWI) is the current standard method to determine final and detailed model parameters to be used in the seismic imaging process. However, FWI is an ill-posed problem that easily achieves a local minimum, leading the model solution in the wrong direction. Recently, some works proposed integrating FWI with Convolutional Neural Networks (CNN). In this case, the CNN weights are updated following the FWI gradient, defining the process as a Physics-Informed Neural Network (PINN). FWI integrated with CNN has an important advantage. The CNN stabilizes the inversion, acting like a regularizer, avoiding local minima-related problems and sparing an initial velocity model in some cases. However, such a process, especially when not requiring an initial model, is computationally expensive due to the high number of iterations required until the convergence. In this work, we propose an approach which relies on combining supervised learning and physics-informed by using a previously trained CNN to start the DL-FWI inversion. Loading the pre-trained weights configures transfer learning. The pre-trained CNN is obtained using a supervised approach based on training with a reduced and simple data set to capture the main velocity trend at the initial FWI iterations. The proposed training process is different from the initial works on the area which obtained the velocity model from the shots in supervised learning tasks and that required a large amount of labelled data to ensure reasonable model predictions. We investigated in our approach two CNN architectures, obtaining more robust results and a reduced number of parameters when using a modified U-Net. The method was probed over three benchmark models, showing consistently that the pre-training phase reduces the process’s uncertainties and accelerates the model convergence using minimal prior information. Besides, the final scores of the iterative process are better than the examples without transfer learning. Thus, transfer learning solved one main limitation of the previous PINN approaches: the unfeasible number of iterations when not using an initial model. Moreover, we tested the method using data with low-frequency band limitations, since the lack of low frequencies is a common issue within real seismic data. The inversion converges to reasonable results probing the method’s robustness with restricted frequency content.

Multi-classification approach for lung nodule detection and classification with proposed texture feature in X-ray images.

Lung cancer is a widespread type of cancer around the world. It is, moreover, a lethal type of tumor. Nevertheless, analysis signifies that earlier recognition of lung cancer considerably develops the possibilities of survival. By deploying X-rays and Computed Tomography (CT) scans, radiologists could identify hazardous nodules at an earlier period. However, when more citizens adopt these diagnoses, the workload rises for radiologists. Computer Assisted Diagnosis (CAD)-based detection systems can identify these nodules automatically and could assist radiologists in reducing their workloads. However, they result in lower sensitivity and a higher count of false positives. The proposed work introduces a new approach for Lung Nodule (LN) detection. At first, Histogram Equalization (HE) is done during pre-processing. As the next step, improved Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) based segmentation is done. Then, the characteristics, including "Gray Level Run-Length Matrix (GLRM), Gray Level Co-Occurrence Matrix (GLCM), and the proposed Local Vector Pattern (LVP)," are retrieved. These features are then categorized utilizing an optimized Convolutional Neural Network (CNN) and itdetectsnodule or non-nodule images. Subsequently, Long Short-Term Memory (LSTM) is deployed to categorize nodule types (benign, malignant, or normal). The CNN weights are fine-tuned by the Chaotic Population-based Beetle Swarm Algorithm (CP-BSA). Finally, the superiority of the proposed approach is confirmed across various measures. The developed approach has exhibited a high precision value of 0.9575 for the best case scenario, and high sensitivity value of 0.9646 for the mean case scenario. The superiority of the proposed approach is confirmed across various measures.

Perceptual Hashing of Deep Convolutional Neural Networks for Model Copy Detection

In recent years, many model intellectual property (IP) proof methods for IP protection have been proposed, such as model watermarking and model fingerprinting. However, with the increasing number of models transmitted and deployed on the Internet, quickly finding the suspect model among thousands of models on model-sharing platforms such as GitHub is in great demand, which concurrently triggers the new security problem of model copy detection for IP protection. As an important part of the model IP protection system, the model copy detection task has not received enough attention. Due to the high computational complexity, both model watermarking and model fingerprinting lack the capability to efficiently find suspected infringing models among tens of millions of models. In this article, inspired by the hash-based image retrieval methods, we introduce a novel model copy detection mechanism: perceptual hashing for convolutional neural networks (CNNs). The proposed perceptual hashing algorithm can convert the weights of CNNs to fixed-length binary hash codes so that the lightly modified version has the similar hash code as the original model. By comparing the similarity of a pair of hash codes between a query model and a test model in the model library, similar versions of a query model can be retrieved efficiently. To the best of our knowledge, this is the first perceptual hashing algorithm for deep neural network models. Specifically, we first select the important model weights based on the model compression theory, then calculate the normal test statistics (NTS) on the segments of important weights, and finally encode the NTS features into hash codes. The experiment performed on a model library containing 3,565 models indicates that our perceptual hashing scheme has a superior copy detection performance.

RBUE: a ReLU-based uncertainty estimation method for convolutional neural networks

Convolutional neural networks (CNNs) have successfully demonstrated their powerful predictive performance in a variety of tasks. However, it remains a challenge to estimate the uncertainty of these predictions simply and accurately. Deep Ensemble is widely considered the state-of-the-art method which can estimate the uncertainty accurately, but it is expensive to train and test. MC-Dropout is another popular method that is less costly but lacks the diversity of predictions resulting in less accurate uncertainty estimates. To combine the benefits of both, we introduce a ReLU-Based Uncertainty Estimation (RBUE) method. Instead of using the randomness of the Dropout module during the test phase (MC-Dropout) or using the randomness of the initial weights of CNNs (Deep Ensemble), RBUE uses the randomness of activation function to obtain diverse outputs in the testing phase to estimate uncertainty. Under the method, we propose strategy MC-DropReLU and develop strategy MC-RReLU. The uniform distribution of the activation function’s position in CNNs allows the randomness to be well transferred to the output results and gives a more diverse output, thus improving the accuracy of the uncertainty estimation. Moreover, our method is simple to implement and does not need to modify the existing model. We experimentally validate the RBUE on three widely used datasets, CIFAR10, CIFAR100, and TinyImageNet. The experiments demonstrate that our method has competitive performance but is more favorable in training time.

Remote Sensing Imagery Object Detection Model Compression via Tucker Decomposition

Although convolutional neural networks (CNNs) have made significant progress, their deployment onboard is still challenging because of their complexity and high processing cost. Tensors provide a natural and compact representation of CNN weights via suitable low-rank approximations. A novel decomposed module called DecomResnet based on Tucker decomposition was proposed to deploy a CNN object detection model on a satellite. We proposed a remote sensing image object detection model compression framework based on low-rank decomposition which consisted of four steps, namely (1) model initialization, (2) initial training, (3) decomposition of the trained model and reconstruction of the decomposed model, and (4) fine-tuning. To validate the performance of the decomposed model in our real mission, we constructed a dataset containing only two classes of objects based on the DOTA and HRSC2016. The proposed method was comprehensively evaluated on the NWPU VHR-10 dataset and the CAST-RS2 dataset created in this work. The experimental results demonstrated that the proposed method, which was based on Resnet-50, could achieve up to 4.44 times the compression ratio and 5.71 times the speedup ratio with merely a 1.9% decrease in the mAP (mean average precision) of the CAST-RS2 dataset and a 5.3% decrease the mAP of the NWPU VHR-10 dataset.

High-level and Low-level Feature Set for Image Caption Generation with Optimized Convolutional Neural Network

Automatic creation of image descriptions, i.e. captioning of images, is an important topic in artificial intelligence (AI) that bridges the gap between computer vision (CV) and natural language processing (NLP). Currently, neural networks are becoming increasingly popular in captioning images and researchers are looking for more efficient models for CV and sequence-sequence systems. This study focuses on a new image caption generation model that is divided into two stages. Initially, low-level features, such as contrast, sharpness, color and their high-level counterparts, such as motion and facial impact score, are extracted. Then, an optimized convolutional neural network (CNN) is harnessed to generate the captions from images. To enhance the accuracy of the process, the weights of CNN are optimally tuned via spider monkey optimization with sine chaotic map evaluation (SMO-SCME). The development of the proposed method is evaluated with a diversity of metrics.

Deep Learning-based SNR Estimation for Multistage Spectrum Sensing in Cognitive Radio Networks

Vacant frequency bands are used in cognitive radio (CR) by incorporating the spectrum sensing (SS) technique. Spectrum sharing plays a central role in ensuring the effectiveness of CR applications. Therefore, a new multi-stage detector for robust signal and spectrum sensing applications is introduced here. Initially, the sampled signal is subjected to SNR estimation by using a convolutional neural network (CNN). Next, the detection strategy is selected in accordance with the predicted SNR levels of the received signal. Energy detector (ED) and singular value-based detector (SVD) are the solutions utilized in the event of high SNR, whilst refined non-negative matrix factorization (MNMF) is employed in the case of low SNR. CNN weights are chosen via the Levy updated sea lion optimization (LU-SLNO) algorithm inspired by the traditional sea lion optimization (SLNO) approach. Finally, the outcomes of the selected detectors are added, offering a precise decision on spectrum tenancy and existence of the signal.

Deep Ensemble Technique for Cyber Attack Detection in Big Data Environment

In this research work, a unique DDoS attack detection and mitigation approach is developed. At first, the input information is carried out into a detection phase, where it identifies the presence of an attack as well as the types of attack. The detection phase will include the “pre-processing, feature extraction, attack detection & attack mitigation.” The collected raw information is pre-processed. Then, the flow-based features such as “flow rate, flow byte, total forward packet, as well as total backward packet” are extracted. In addition, the sequential frequent pattern characteristics are retrieved utilizing the Apriori model. For the precise detection, an ensemble classifier is introduced by inclosing the “Neural Network (ANN), Bi-GRU, Recurrent Neural Network (RNN), and optimized Convolutional Neural Network (CNN).” The DBN, Bi-GRU, and RNN of ensemble classifiers are trained with the retrieved characteristics. The result from DBN, Bi-GRU, and RNN is subjected as an input to an optimized CNN, in which weights of CNN are optimally tuned by a novel hybrid optimization. Then, the Trust_BAIT strategy is used in the mitigation phase to remove the intruder from the network.

Multiarea Inertia Estimation Using Convolutional Neural Networks and Federated Learning

With the increase in penetration of renewable energy sources (RES), traditional inertia estimation techniques based purely on the number of online synchronous generators are increasingly unsuitable, ultimately leading towards suboptimal frequency control in the electric power grid. The stochastic nature of RES additionally makes the system inertia a time-varying quantity. Furthermore, the frequency and inertial response of power systems change drastically in multiarea power systems with interconnected tie-lines. Hence, it is important for state/parameter estimation (e.g., inertia) in multiarea systems, while ensuring communication between each of the areas. In this article, a client–server-based federated learning framework is used to estimate power system inertia in a multiarea system. Federated learning is a machine learning technique where multiple decentralized devices are trained with local data, and a global model is updated and redistributed by a central server by aggregating the trained weights of the decentralized devices, without exchanging the local data. Using local frequency measurements, obtained from the phase-locked loop of an energy storage system, the inertia at each of the areas can be estimated locally via offline training using convolutional neural networks (CNNs), whereas the CNN weights update in an online fashion. The framework, tested on a two-area power system, accurately estimated the inertia constant for both independent and identically distributed (IID) and non-IID data. Furthermore, the CNN-based method outperformed conventional neural network-based estimation techniques in terms of number of communication rounds and estimation accuracy.

ExplainFix : Explainable spatially fixed deep networks

Is there an initialization for deep networks that requires no learning? ExplainFix adopts two design principles: the “fixed filters” principle that all spatial filter weights of convolutional neural networks can be fixed at initialization and never learned, and the “nimbleness” principle that only few network parameters suffice. We contribute (a) visual model-based explanations, (b) speed and accuracy gains, and (c) novel tools for deep convolutional neural networks. ExplainFix gives key insights that spatially fixed networks should have a steered initialization, that spatial convolution layers tend to prioritize low frequencies, and that most network parameters are not necessary in spatially fixed models. ExplainFix models have up to ×100 fewer spatial filter kernels than fully learned models and matching or improved accuracy. Our extensive empirical analysis confirms that ExplainFix guarantees nimbler models (train up to 17% faster with channel pruning), matching or improved predictive performance (spanning 13 distinct baseline models, four architectures and two medical image datasets), improved robustness to larger learning rate, and robustness to varying model size. We are first to demonstrate that all spatial filters in state-of-the-art convolutional deep networks can be fixed at initialization, not learned. This article is categorized under: Technologies > Machine Learning Fundamental Concepts of Data and Knowledge > Explainable AI Fundamental Concepts of Data and Knowledge > Key Design Issues in Data Mining

Direct Evaluation of Treatment Response in Brain Metastatic Disease with Deep Neuroevolution.

Cancer centers have an urgent and unmet clinical and research need for AI that can guide patient management. A core component of advancing cancer treatment research is assessing response to therapy. Doing so by hand, for example, as per RECIST or RANO criteria, is tedious and time-consuming, and can miss important tumor response information. Most notably, the prevalent response criteria often exclude lesions, the non-target lesions, altogether. We wish to assess change in a holistic fashion that includes all lesions, obtaining simple, informative, and automated assessments of tumor progression or regression. Because genetic sub-types of cancer can be fairly specific and patient enrollment in therapy trials is often limited in number and accrual rate, we wish to make response assessments with small training sets. Deep neuroevolution (DNE) is a novel radiology artificial intelligence (AI) optimization approach that performs well on small training sets. Here, we use a DNE parameter search to optimize a convolutional neural network (CNN) that predicts progression versus regression of metastatic brain disease. We analyzed 50 pairs of MRI contrast-enhanced images as our training set. Half of these pairs, separated in time, qualified as disease progression, while the other 25 image pairs constituted regression. We trained the parameters of a CNN via "mutations" that consisted of random CNN weight adjustments and evaluated mutation "fitness" as summed training set accuracy. We then incorporated the best mutations into the next generation's CNN, repeating this process for approximately 50,000 generations. We applied the CNNs to our training set, as well as a separate testing set with the same class balance of 25 progression and 25 regression cases. DNE achieved monotonic convergence to 100% training set accuracy. DNE also converged monotonically to 100% testing set accuracy. We have thus shown thatDNE can accurately classify brain metastatic disease progression versus regression. Future work will extend the input from 2D image slices to full 3D volumes, and include the category of "no change." We believe that an approach such as ours can ultimately provide a useful and informative complement to RANO/RECIST assessment and volumetric AI analysis.

Deep Bayesian inference for seismic imaging with tasks

We use techniques from Bayesian inference and deep neural networks to translate uncertainty in seismic imaging to uncertainty in tasks performed on the image, such as horizon tracking. Seismic imaging is an ill-posed inverse problem because of bandwidth and aperture limitations, which are hampered by the presence of noise and linearization errors. Many regularization methods, such as transform-domain sparsity promotion, have been designed to deal with the adverse effects of these errors; however, these methods run the risk of biasing the solution and do not provide information on uncertainty in the image space and how this uncertainty impacts certain tasks on the image. A systematic approach is developed to translate uncertainty due to noise in the data to the confidence intervals of automatically tracked horizons in the image. The uncertainty in the seismic image is characterized by a convolutional neural network (CNN) that is used to reparameterize the image. To assess these uncertainties, samples are drawn from the posterior distribution of the CNN weights. Compared with traditional priors, it is argued in the literature that these CNNs introduce a flexible inductive bias that is a surprisingly good fit for a diverse set of problems, including medical imaging, compressive sensing, and diffraction tomography. The method of stochastic gradient Langevin dynamics is used to sample from the posterior distribution. This method is designed to handle large-scale Bayesian inference problems with computationally expensive forward operators as in seismic imaging. Aside from offering a robust alternative to the maximum a posteriori estimate that is prone to overfitting, access to these samples allows us to translate uncertainty in the image, due to noise in the data, to uncertainty on the tracked horizons. For instance, it admits estimates for the pointwise standard deviation on the image and for confidence intervals on its automatically tracked horizons.

Video-based emotion sensing and recognition using convolutional neural network based kinetic gas molecule optimization

Human facial expressions are thought to be important in interpreting one's emotions. Emotional recognition plays a very important part in the more exact inspection of human feelings and interior thoughts. Over the last several years, emotion identification utilizing pictures, videos, or voice as input has been a popular issue in the field of study. Recently, most emotional recognition research focuses on the extraction of representative modality characteristics and the definition of dynamic interactions between multiple modalities. Deep learning methods have opened the way for the development of artificial intelligence products, and the suggested system employs a convolutional neural network (CNN) for identifying real-time human feelings. The aim of the research study is to create a real-time emotion detection application by utilizing improved CNN. This research offers information on identifying emotions in films using deep learning techniques. Kinetic gas molecule optimization is used to optimize the fine-tuning and weights of CNN. This article describes the technique of the recognition process as well as its experimental validation. Two datasets such as video-based and image-based datasets, which are employed in many scholarly publications, are also investigated. The results of several emotion recognition simulations are provided, along with their performance factors.

VWC-SDK: Convolutional Weight Mapping Using Shifted and Duplicated Kernel With Variable Windows and Channels

With their high energy efficiency, processing-in-memory (PIM) arrays are being increasingly used for the convolutional neural network (CNN) inference. In the PIM-based CNN inference, the computational latency and energy are dependent on how the CNN weights are mapped to the PIM array. A recent study proposed the shifted and duplicated kernel (SDK) mapping that reuses input feature maps with a unit of a parallel window, which is convolved with duplicated kernels to obtain multiple output elements in parallel. However, the existing SDK-based mapping algorithm does not always result in minimum computing cycles because it only maps a square-shaped parallel window with the entire channels. In this paper, we introduce a novel mapping algorithm called variable-window SDK (VW-SDK), which adaptively determines the shape of the parallel window that leads to the minimum computing cycles for the given convolutional layer and PIM array. By allowing rectangular-shaped windows with partial channels, VW-SDK utilizes the PIM array more efficiently, thereby further reducing the number of computing cycles. To further remove the inefficient computing cycles caused by the residual channels, we extend the VW-SDK algorithm into VWC-SDK (SDK with variable windows and channels) that additionally performs residual channel pruning. The simulation with a <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$512\times 512$ </tex-math></inline-formula> PIM array and ResNet-20 shows that VW-SDK improves the inference speed by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.29\times $ </tex-math></inline-formula> compared to the existing SDK-based algorithm. The results also show that residual channel pruning improves the inference speed of ResNet-20 by up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 1.38\times $ </tex-math></inline-formula> when compared to the original network without pruning.