Gokul Yenduri
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Gokul Yenduri has a total of 65 co-authors. Here are the top co-authors by number of shared publications:
- Thippa Reddy Gadekallu
Lovely Professional University
14 shared publications - Praveen Kumar Reddy Maddikunta
School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamil Nadu, India
9 shared publications - Gautam Srivastava
Sogang University, Seoul, Republic of Korea
6 shared publications - Nancy Victor
School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, India
6 shared publications - Sweta Bhattacharya
School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamil Nadu, India
5 shared publications - Dasaradharami Reddy Kandati
Vellore Institute of Technology
4 shared publications
Latest Publications
- Open Access
- 2 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 33237-33251
Privacy-Preserving Multi-Class Skin Lesion Classification Using Single-Shot Machine Unlearning and Noise Maximization
The fusion of data analytics, machine learning, and artificial intelligence has transformed various sectors, notably healthcare. This rapid advancement has also revolutionized dermatology, offering remarkable accuracy in skin lesion classification. In highly confidential domains like dermatology, the task of unlearning the model’s training data becomes crucial. Unlearning optimizes data processing by deleting unnecessary information, thereby lowering the burden on healthcare devices. However, the integration of these models into clinical environments presents a unique challenge: the need to balance model performance with patient privacy. This paper presents a novel approach addressing the imperative task of unlearning specific classes from a pretrained model, focusing on pigmented skin lesions. We present a one-epoch method that adds noise to trained models using the HAM10000 dataset, strategically targeting the forgetting of specific classes. Then, we use a fine-tuning phase with a single epoch to re-calibrate the model for the remaining classes. This provides a full framework for perturbation and recovery. We evaluate the efficiency of this approach across four distinct model architectures: MobileNet-V2, EfficientNetB2, ResNet-50, and VGG-19. We demonstrate the robustness and adaptability of the proposed method, providing insights into its applicability in real-world clinical environment for dynamic data management and model refinement.
- 2 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 33237-33251
Open AccessPrivacy-Preserving Multi-Class Skin Lesion Classification Using Single-Shot Machine Unlearning and Noise Maximization
The fusion of data analytics, machine learning, and artificial intelligence has transformed various sectors, notably healthcare. This rapid advancement has also revolutionized dermatology, offering remarkable accuracy in skin lesion classification. In highly confidential domains like dermatology, the task of unlearning the model’s training data becomes crucial. Unlearning optimizes data processing by deleting unnecessary information, thereby lowering the burden on healthcare devices. However, the integration of these models into clinical environments presents a unique challenge: the need to balance model performance with patient privacy. This paper presents a novel approach addressing the imperative task of unlearning specific classes from a pretrained model, focusing on pigmented skin lesions. We present a one-epoch method that adds noise to trained models using the HAM10000 dataset, strategically targeting the forgetting of specific classes. Then, we use a fine-tuning phase with a single epoch to re-calibrate the model for the remaining classes. This provides a full framework for perturbation and recovery. We evaluate the efficiency of this approach across four distinct model architectures: MobileNet-V2, EfficientNetB2, ResNet-50, and VGG-19. We demonstrate the robustness and adaptability of the proposed method, providing insights into its applicability in real-world clinical environment for dynamic data management and model refinement.
- Open Access
- 25 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 134325-134356
Artificial General Intelligence: Advancements, Challenges, and Future Directions in AGI Research
Artificial General Intelligence (AGI) is a transformative shift in artificial intelligence that aims to match human-like reasoning, flexibility, and self-learning. Unlike Narrow AI, which is capable of performing only a limited set of tasks, AGI aspires to handle any intellectual task by exhibiting human-like learning, reasoning, and behavior. This enables AGI to offer extraordinary potential in various domains such as healthcare, education, transportation, and more. This paper provides a comprehensive review of the fundamental concepts, applications, and challenges associated with AGI’s development. This systematic review of the literature explored AGI’s transformative potential, from personalized healthcare and adaptive learning systems to advanced autonomous vehicles and predictive analytics. Although, AGI has great potential, multiple challenges, such as ethical issues, data privacy, and other technical challenges, must be addressed before its launch in the real world.
- 25 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 134325-134356
Open AccessArtificial General Intelligence: Advancements, Challenges, and Future Directions in AGI Research
Artificial General Intelligence (AGI) is a transformative shift in artificial intelligence that aims to match human-like reasoning, flexibility, and self-learning. Unlike Narrow AI, which is capable of performing only a limited set of tasks, AGI aspires to handle any intellectual task by exhibiting human-like learning, reasoning, and behavior. This enables AGI to offer extraordinary potential in various domains such as healthcare, education, transportation, and more. This paper provides a comprehensive review of the fundamental concepts, applications, and challenges associated with AGI’s development. This systematic review of the literature explored AGI’s transformative potential, from personalized healthcare and adaptive learning systems to advanced autonomous vehicles and predictive analytics. Although, AGI has great potential, multiple challenges, such as ethical issues, data privacy, and other technical challenges, must be addressed before its launch in the real world.
- Open Access
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Computer Society
- Vol. 6, 4-24
The Metaverse for Industry 5.0 in NextG Communications: Potential Applications and Future Challenges
With the advent of new technologies and endeavours for automation in almost all day-to-day activities, the recent discussions on the metaverse life have a greater expectation. The metaverse enables people to communicate with each other by combining the physical world with the virtual world. However, realizing the Metaverse requires symmetric content delivery, low latency, dynamic network control, etc. Industry 5.0 is expected to reform the manufacturing processes through human-robot collaboration and effective utilization of technologies like Artificial intelligence for increased productivity and less maintenance. The metaverse with Industry 5.0 may have tremendous technological integration for a more immersive experience and enhanced productivity. In this review, we present an overview of the metaverse and Industry 5.0, focusing on key technologies that enable the industrial metaverse, including virtual and augmented reality, 3D modeling, artificial intelligence, edge computing, digital twins, blockchain, and 6G communication networks. The paper then discusses the metaverse's diverse applications across various Industry 5.0 sectors, such as agriculture, supply chain management, healthcare, education, and transportation, illustrated through several research initiatives. Additionally, the paper addresses the challenges of implementing the industrial metaverse, proposes potential solutions, and outlines directions for future research.
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Computer Society
- Vol. 6, 4-24
Open AccessThe Metaverse for Industry 5.0 in NextG Communications: Potential Applications and Future Challenges
With the advent of new technologies and endeavours for automation in almost all day-to-day activities, the recent discussions on the metaverse life have a greater expectation. The metaverse enables people to communicate with each other by combining the physical world with the virtual world. However, realizing the Metaverse requires symmetric content delivery, low latency, dynamic network control, etc. Industry 5.0 is expected to reform the manufacturing processes through human-robot collaboration and effective utilization of technologies like Artificial intelligence for increased productivity and less maintenance. The metaverse with Industry 5.0 may have tremendous technological integration for a more immersive experience and enhanced productivity. In this review, we present an overview of the metaverse and Industry 5.0, focusing on key technologies that enable the industrial metaverse, including virtual and augmented reality, 3D modeling, artificial intelligence, edge computing, digital twins, blockchain, and 6G communication networks. The paper then discusses the metaverse's diverse applications across various Industry 5.0 sectors, such as agriculture, supply chain management, healthcare, education, and transportation, illustrated through several research initiatives. Additionally, the paper addresses the challenges of implementing the industrial metaverse, proposes potential solutions, and outlines directions for future research.
- 25 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IT Professional
- Vol. 26  (4) , 22-28
Enhancing Connectivity in Rural Areas: Secure Spectrum Access in 6G Networks Using Advanced Encryption and Spectrum Sensing Techniques
The advancement of 6G cognitive radio networks aims to reduce latency in rural and remote areas. Very few studies have been conducted on this technology. Therefore, this study utilizes massive multiple-input, multiple-output (MIMO) technology for secure data transmission at 6G base stations. Blockchain technology authenticates IDs and maintains secure records for network users, with decentralization achieved through the chimp optimization algorithm. The availability of the spectrum is monitored using the Q-learning hidden sparse variate logistic regression model, and the channel-state information is predicted using the quasi-Newton iterative unscented Kalman filter algorithm. Additionally, beamforming is enhanced through cooperative strategies. Secure routing is facilitated by the golden eagle optimization-hyper elliptic curve cryptography algorithm, where data are routed according to paths determined by the Dijkstra algorithm. The MIMO-6G-cognitive radio-based Internet-of-Things framework performs better compared to existing methods.
- 25 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IT Professional
- Vol. 26  (4) , 22-28
Enhancing Connectivity in Rural Areas: Secure Spectrum Access in 6G Networks Using Advanced Encryption and Spectrum Sensing Techniques
The advancement of 6G cognitive radio networks aims to reduce latency in rural and remote areas. Very few studies have been conducted on this technology. Therefore, this study utilizes massive multiple-input, multiple-output (MIMO) technology for secure data transmission at 6G base stations. Blockchain technology authenticates IDs and maintains secure records for network users, with decentralization achieved through the chimp optimization algorithm. The availability of the spectrum is monitored using the Q-learning hidden sparse variate logistic regression model, and the channel-state information is predicted using the quasi-Newton iterative unscented Kalman filter algorithm. Additionally, beamforming is enhanced through cooperative strategies. Secure routing is facilitated by the golden eagle optimization-hyper elliptic curve cryptography algorithm, where data are routed according to paths determined by the Dijkstra algorithm. The MIMO-6G-cognitive radio-based Internet-of-Things framework performs better compared to existing methods.
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11  (23) , 37708-37717
A Trustable Federated Learning Framework for Rapid Fire Smoke Detection at the Edge in Smart Home Environments
With the rapid growth of the Internet of Things, sensors have become integral components of smart homes, enabling real-time monitoring and control of various aspects ranging from energy consumption to security. In this context, we cannot underestimate the importance of sensor-based data in ensuring the safety and well-being of occupants, particularly in scenarios involving early detection of fire outbreaks. We propose a novel Federated Learning (FL) Framework in this study to address the crucial issue of rapid fire smoke detection at the edge of smart home environments. The proposed framework employs three distinct FL algorithms, namely Federated Averaging, Federated Adaptive Moment Estimation, and Federated Proximal, for global aggregation of machine learning predictions based on data from various IoT sensors. This framework allows for early prediction by utilizing the computational capabilities at the edge, thereby improving the responsiveness and efficiency of fire safety systems. Furthermore, to improve trust and transparency in the FL framework, eXplainable Artificial Intelligence techniques such as Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive exPlanations (SHAP) are integrated. We unveil pivotal features driving predictive outcomes through LIME and SHAP analyses, offering users valuable insights into model decision-making processes.
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11  (23) , 37708-37717
A Trustable Federated Learning Framework for Rapid Fire Smoke Detection at the Edge in Smart Home Environments
With the rapid growth of the Internet of Things, sensors have become integral components of smart homes, enabling real-time monitoring and control of various aspects ranging from energy consumption to security. In this context, we cannot underestimate the importance of sensor-based data in ensuring the safety and well-being of occupants, particularly in scenarios involving early detection of fire outbreaks. We propose a novel Federated Learning (FL) Framework in this study to address the crucial issue of rapid fire smoke detection at the edge of smart home environments. The proposed framework employs three distinct FL algorithms, namely Federated Averaging, Federated Adaptive Moment Estimation, and Federated Proximal, for global aggregation of machine learning predictions based on data from various IoT sensors. This framework allows for early prediction by utilizing the computational capabilities at the edge, thereby improving the responsiveness and efficiency of fire safety systems. Furthermore, to improve trust and transparency in the FL framework, eXplainable Artificial Intelligence techniques such as Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive exPlanations (SHAP) are integrated. We unveil pivotal features driving predictive outcomes through LIME and SHAP analyses, offering users valuable insights into model decision-making processes.
- Open Access
- 8 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 116987-117007
Demystifying Defects: Federated Learning and Explainable AI for Semiconductor Fault Detection
Semiconductor manufacturing, a critical driver of modern technology, involves intricate processes for fabricating integrated circuits on materials like silicon. This industry’s pivotal role spans various applications, from smartphones to computers, emphasizing the importance of fault detection to ensure the reliability and cost-efficiency of electronic devices. Fault detection within this sector entails collaboration among multiple stakeholders, including Original Equipment Manufacturers (OEMs), Integrated Device Manufacturers (IDMs), wafer foundries, and software providers. A common challenge is the reluctance to share sensitive design data centrally, which is essential for building traditional machine learning models. To overcome these challenges, this paper introduces an innovative fault detection model that leverages Federated Learning (FL) and Explainable AI (XAI). FL’s decentralized approach enhances model learning across multiple nodes without requiring the pooling of sensitive data, thus preserving data privacy. Concurrently, XAI ensures that the developed models maintain transparency and trustworthiness, even when trained on distributed datasets. This FL-based fault detection model permits stakeholders to train ML models on node-specific data without centralizing sensitive information. It accommodates heterogeneous and asynchronously-stored data, diverse machine learning models, and nodes with varying capacities and data volumes. By addressing the opacity of deep learning models, FL and XAI unveil their predictive behaviors in identifying semiconductor faults. Empirical results, obtained using a public dataset, demonstrate a significant improvement in defect identification precision, achieving an exceptional test accuracy of 98.78%. These findings underscore the potential of the proposed approach to transform fault detection in semiconductor manufacturing, thereby enhancing the reliability and efficiency of the production process.
- 8 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 116987-117007
Open AccessDemystifying Defects: Federated Learning and Explainable AI for Semiconductor Fault Detection
Semiconductor manufacturing, a critical driver of modern technology, involves intricate processes for fabricating integrated circuits on materials like silicon. This industry’s pivotal role spans various applications, from smartphones to computers, emphasizing the importance of fault detection to ensure the reliability and cost-efficiency of electronic devices. Fault detection within this sector entails collaboration among multiple stakeholders, including Original Equipment Manufacturers (OEMs), Integrated Device Manufacturers (IDMs), wafer foundries, and software providers. A common challenge is the reluctance to share sensitive design data centrally, which is essential for building traditional machine learning models. To overcome these challenges, this paper introduces an innovative fault detection model that leverages Federated Learning (FL) and Explainable AI (XAI). FL’s decentralized approach enhances model learning across multiple nodes without requiring the pooling of sensitive data, thus preserving data privacy. Concurrently, XAI ensures that the developed models maintain transparency and trustworthiness, even when trained on distributed datasets. This FL-based fault detection model permits stakeholders to train ML models on node-specific data without centralizing sensitive information. It accommodates heterogeneous and asynchronously-stored data, diverse machine learning models, and nodes with varying capacities and data volumes. By addressing the opacity of deep learning models, FL and XAI unveil their predictive behaviors in identifying semiconductor faults. Empirical results, obtained using a public dataset, demonstrate a significant improvement in defect identification precision, achieving an exceptional test accuracy of 98.78%. These findings underscore the potential of the proposed approach to transform fault detection in semiconductor manufacturing, thereby enhancing the reliability and efficiency of the production process.
- Open Access
- 21 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 87131-87151
Handling the Class Imbalance Problem With an Improved Sine Cosine Algorithm for Optimal Instance Selection
Class imbalance is a significant study problem that is biased, exhibiting excellent performance toward the majority classes in the dataset while showing inferior performance toward minority classes. When dealing with real-world issues, this kind of biased nature affects classification accuracy. The Improved Binary Sine Cosine Algorithm (IBSCA) has been used in this work to identify a subset of the majority class in the best possible way. The proposed IBSCA makes some enhancements over the conventional Binary Sine Cosine Algorithm (BSCA) to address the issue of premature convergence with local optimal solutions. To improve classification accuracy for unbalanced datasets, the proposed IBSCA seeks to identify the optimal collection of instances from the majority class. The advised IBSCA makes use of the alpha agent, beta agent, and random agent’s location, which tends to devote considerable time to exploration to find the best possible set of instances. By using the geometric mean (G-mean) and F-score to describe the fitness function, the proposed IBSCA aims to solve the multi-objective optimization issue. On 18 datasets with different imbalance ratios taken from the KEEL repository, experimentation is conducted. Comparisons are made between the suggested IBSCA and the traditional Binary Sine Cosine Algorithm, Binary Particle Swarm Optimization (BPSO), and Binary GreyWolf Optimization (BGWO). Additionally, the performance of the suggested IBSCA is evaluated against the top outcomes from different research papers. Metrics like sensitivity, F-score, G-mean, and area under curve (AUC) show that the suggested IBSCA outperforms the state-of-the-art algorithms. The statistical findings using the Wilcoxon signed rank test and Friedman test also demonstrate that the suggested IBSCA is more efficient than the other conventional algorithms.
- 21 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 87131-87151
Open AccessHandling the Class Imbalance Problem With an Improved Sine Cosine Algorithm for Optimal Instance Selection
Class imbalance is a significant study problem that is biased, exhibiting excellent performance toward the majority classes in the dataset while showing inferior performance toward minority classes. When dealing with real-world issues, this kind of biased nature affects classification accuracy. The Improved Binary Sine Cosine Algorithm (IBSCA) has been used in this work to identify a subset of the majority class in the best possible way. The proposed IBSCA makes some enhancements over the conventional Binary Sine Cosine Algorithm (BSCA) to address the issue of premature convergence with local optimal solutions. To improve classification accuracy for unbalanced datasets, the proposed IBSCA seeks to identify the optimal collection of instances from the majority class. The advised IBSCA makes use of the alpha agent, beta agent, and random agent’s location, which tends to devote considerable time to exploration to find the best possible set of instances. By using the geometric mean (G-mean) and F-score to describe the fitness function, the proposed IBSCA aims to solve the multi-objective optimization issue. On 18 datasets with different imbalance ratios taken from the KEEL repository, experimentation is conducted. Comparisons are made between the suggested IBSCA and the traditional Binary Sine Cosine Algorithm, Binary Particle Swarm Optimization (BPSO), and Binary GreyWolf Optimization (BGWO). Additionally, the performance of the suggested IBSCA is evaluated against the top outcomes from different research papers. Metrics like sensitivity, F-score, G-mean, and area under curve (AUC) show that the suggested IBSCA outperforms the state-of-the-art algorithms. The statistical findings using the Wilcoxon signed rank test and Friedman test also demonstrate that the suggested IBSCA is more efficient than the other conventional algorithms.
- Open Access
- 4 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 36764-36777
Enhancing Household Energy Consumption Predictions Through Explainable AI Frameworks
Effective energy management is crucial for sustainability, carbon reduction, resource conservation, and cost savings. However, conventional energy forecasting methods often lack accuracy, suggesting the need for advanced approaches. Artificial intelligence (AI) has emerged as a powerful tool for energy forecasting, but its lack of transparency and interpretability poses challenges for understanding its predictions. In response, Explainable AI (XAI) frameworks have been developed to enhance the transparency and interpretability of black-box AI models. Accordingly, this paper focuses on achieving accurate household energy consumption predictions by comparing prediction models based on several evaluation metrics, namely the Coefficient of Determination (R 2 ), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE). The best model is identified by comparison after making predictions on unseen data, after which the predictions are explained by leveraging two XAI frameworks: Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP). These explanations help identify crucial characteristics contributing to energy consumption predictions, including insights into feature importance. Our findings underscore the significance of current consumption patterns and lagged energy consumption values in estimating energy usage. This paper further demonstrates the role of XAI in developing consistent and reliable predictive models.
- 4 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 36764-36777
Open AccessEnhancing Household Energy Consumption Predictions Through Explainable AI Frameworks
Effective energy management is crucial for sustainability, carbon reduction, resource conservation, and cost savings. However, conventional energy forecasting methods often lack accuracy, suggesting the need for advanced approaches. Artificial intelligence (AI) has emerged as a powerful tool for energy forecasting, but its lack of transparency and interpretability poses challenges for understanding its predictions. In response, Explainable AI (XAI) frameworks have been developed to enhance the transparency and interpretability of black-box AI models. Accordingly, this paper focuses on achieving accurate household energy consumption predictions by comparing prediction models based on several evaluation metrics, namely the Coefficient of Determination (R 2 ), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE). The best model is identified by comparison after making predictions on unseen data, after which the predictions are explained by leveraging two XAI frameworks: Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP). These explanations help identify crucial characteristics contributing to energy consumption predictions, including insights into feature importance. Our findings underscore the significance of current consumption patterns and lagged energy consumption values in estimating energy usage. This paper further demonstrates the role of XAI in developing consistent and reliable predictive models.
- 26 November 2023
- Published by Springer Nature
- p. 413-424
PSO-Enabled Federated Learning for Detecting Ships in Supply Chain Management
- 26 November 2023
- Published by Springer Nature
- p. 413-424
PSO-Enabled Federated Learning for Detecting Ships in Supply Chain Management