Postdoc Researcher at Zhongshan Institute of Changchun University of Science and Technology, China.
Dr. Aakash Kumar is a dedicated researcher in control science and engineering, with expertise in deep learning, machine learning, and artificial intelligence applications. He is currently a Postdoctoral Researcher at Zhongshan Institute of Changchun University of Science and Technology in China. His work focuses on developing computational techniques to optimize deep neural networks for image analysis and robotic systems. Throughout his career, Dr. Kumar has contributed to cutting-edge research in AI-driven fault detection, spiking neural networks, and generative models. Fluent in English, Chinese, Urdu, and Sindhi, he has built an international academic and professional profile.
Professional Profile:
Scopus
Orcid
Google Scholar
Education Background
Dr. Kumar earned his Doctor of Engineering in Control Science and Engineering from the University of Science and Technology of China (USTC) in 2022. His research was fully funded by the Chinese Academy of Sciences-The World Academy of Sciences President’s Fellowship. Prior to this, he obtained his Master of Engineering in Control Science and Engineering from USTC in 2017 under the Chinese Government Scholarship. He also completed a Diploma in Chinese Language (HSK-4 Level) at Anhui Normal University in 2014. His academic journey began with a Bachelor of Science in Electronic Engineering from the University of Sindh, Jamshoro, Pakistan, in 2011.
Professional Development
Since 2022, Dr. Kumar has been serving as a Postdoctoral Researcher at Zhongshan Institute of Changchun University of Science and Technology, where he is engaged in pioneering work on deep learning applications, computational intelligence, and machine learning-based fault detection. Prior to this, he worked remotely as a Machine Learning Engineer at COSIMA.AI Inc., New York, where he developed AI models for healthcare, computer vision, and smart systems. His early career included roles as a Data Scientist at Japan Cooperation Agency in Pakistan (2012–2013), where he analyzed agricultural and livestock data using statistical tools, and as a Lecturer at The Pioneers College, Jamshoro (2011–2012).
Research Focus
Dr. Kumar’s research focuses on the optimization of deep neural networks, reinforcement learning, and computational intelligence. His notable projects include the development of a Deep Spiking Q-Network (DSQN) for mobile robot path planning, a CNN-LSTM-AM framework for UAV fault detection, and a Deep Conditional Generative Model for Dictionary Learning (DCGMDL) to enhance classification efficiency. His interests extend to collaborative data analysis, regression modeling, clustering techniques, and Bayesian networks. He is also actively guiding research scholars, including two Ph.D. candidates and a master’s student.
Author Metrics:
Dr. Kumar has presented his research at prestigious conferences, including the International Symposium of Space Optical Instrument and Application in Beijing and academic meetings at USTC. His work on generative AI, deep learning, and autonomous systems has been recognized in academic circles. He has also served as a reviewer for reputed journals such as Neural Processing Letters, Journal of Machine Learning and Cybernetics, The Big Data, and Neural Computing and Applications, all published by Springer. His contributions to AI research and computational intelligence have garnered citations, reflecting his impact in the field.
Honors & Awards
Dr. Kumar has received multiple prestigious scholarships and fellowships, including the Chinese Academy of Sciences-The World Academy of Sciences President’s Fellowship for his Ph.D. and the Chinese Government Scholarship for both his master’s degree and language studies. He has been recognized for his contributions to AI and deep learning applications in autonomous systems, earning invitations to present his work at international conferences. Additionally, his innovative projects in AI-driven fault detection and predictive modeling have gained recognition in the research community.
Publication Top Notes
1. Pruning filters with L1-norm and capped L1-norm for CNN compression
- Authors: A Kumar, AM Shaikh, Y Li, H Bilal, B Yin
- Journal: Applied Intelligence
- Volume: 51, Pages: 1152-1160
- Citations: 144 (2021)
- Key Contribution:
- Introduced an L1-norm and capped L1-norm-based pruning method for CNN model compression.
- Reduced redundant filters, leading to efficient deep learning models with lower computational cost and minimal performance degradation.
2. Jerk-bounded trajectory planning for rotary flexible joint manipulator: an experimental approach
- Authors: H Bilal, B Yin, A Kumar, M Ali, J Zhang, J Yao
- Journal: Soft Computing
- Volume: 27 (7), Pages: 4029-4039
- Citations: 115 (2023)
- Key Contribution:
- Developed a jerk-bounded trajectory planning method to improve the performance of a rotary flexible joint manipulator.
- Conducted experimental validation, proving improved stability and accuracy in robotic movement.
3. Real-time lane detection and tracking for advanced driver assistance systems
- Authors: H Bilal, B Yin, J Khan, L Wang, J Zhang, A Kumar
- Conference: 2019 Chinese Control Conference (CCC)
- Pages: 6772-6777
- Citations: 99 (2019)
- Key Contribution:
- Proposed a real-time lane detection and tracking system for ADAS (Advanced Driver Assistance Systems).
- Used computer vision and deep learning to enhance road safety and autonomous driving technologies.
4. Reduction of multiplications in convolutional neural networks
- Authors: M Ali, B Yin, A Kumar, AM Sheikh, H Bilal
- Conference: 2020 39th Chinese Control Conference (CCC)
- Pages: 7406-7411
- Citations: 85 (2020)
- Key Contribution:
- Developed a method to reduce the number of multiplications in CNN computations, improving efficiency.
- Aimed at hardware acceleration for deep learning models.
5. Using feature entropy to guide filter pruning for efficient convolutional networks
- Authors: Y Li, L Wang, S Peng, A Kumar, B Yin
- Conference: Artificial Neural Networks and Machine Learning – ICANN 2019: Deep Learning
- Citations: 16 (2019)
- Key Contribution:
- Introduced feature entropy-based filter pruning to optimize CNN performance while maintaining accuracy.
- Focused on reducing computational complexity in deep learning applications.
Conclusion
Dr. Aakash Kumar is an exceptional candidate for the Best Researcher Award due to his strong publication record, impactful AI research, interdisciplinary contributions, and academic leadership. His high citation count, expertise in CNN compression, deep learning efficiency, and AI-driven fault detection, along with his postdoctoral research at a leading Chinese university, make him a compelling nominee.
To further strengthen his candidacy, expanding into patents, industry applications, and first-author publications in top AI journals would enhance his global research impact.