Talk to our Students Apply Now Convocation 2026

Expert Talk on Compression of Deep Neural Network Using Genetic Algorithms

Event Date: 06th April 2026

Event brief description

The Department of Artificial Intelligence, School of Artificial Intelligence (SOAI), Galgotias University successfully organized an expert talk on “Compression and Acceleration of Deep Neural Networks Using Genetic Algorithms” on 6th April 2026 at 10:00 AM. The session was delivered by Mr. Suneet Gupta, an experienced professional in the field of Artificial Intelligence and optimization techniques.

The talk focused on addressing one of the key challenges in modern AI systems—efficient deployment of deep neural networks. Mr. Gupta provided valuable insights into how genetic algorithms can be leveraged to optimize model architectures, reduce computational complexity, and enhance performance without significantly compromising accuracy. He explained real-world applications where model compression and acceleration play a crucial role, particularly in edge devices and resource-constrained environments.

The session was highly interactive, encouraging students and faculty members to engage in discussions on advanced AI techniques and research directions. Participants gained a deeper understanding of evolutionary algorithms and their integration with deep learning models.

Overall, the expert talk was informative and inspiring, contributing to the academic and research ecosystem of the university while motivating students to explore innovative solutions in AI and machine learning.

Event Detailed Description

The Department of Artificial Intelligence, School of Artificial Intelligence (SOAI), Galgotias University, organized an insightful expert talk on “Compression and Acceleration of Deep Neural Networks Using Genetic Algorithms” on 6th April 2026 at 10:00 AM. The session was delivered by Mr. Suneet Gupta, a distinguished expert known for his contributions to artificial intelligence, optimization techniques, and efficient model deployment.

The primary objective of the session was to address the growing challenges associated with deploying deep neural networks in real-world environments, particularly where computational resources are limited. Mr. Gupta began the session by introducing the fundamentals of deep neural networks and the need for model optimization. He emphasized that while deep learning models are becoming increasingly powerful, their size and computational requirements often make them difficult to deploy efficiently, especially on edge devices such as mobile phones, IoT systems, and embedded platforms.

The speaker then introduced the concept of genetic algorithms as an evolutionary optimization technique inspired by natural selection. He explained how these algorithms can be effectively applied to neural network pruning, weight optimization, and architecture search. Through practical examples and case studies, Mr. Gupta demonstrated how genetic algorithms help in reducing model size, improving inference speed, and maintaining accuracy, thereby achieving an optimal balance between performance and efficiency.

A key highlight of the session was the discussion on real-world applications, including smart healthcare systems, autonomous vehicles, and intelligent surveillance, where efficient AI models are critical. The talk also covered emerging research trends, such as hybrid optimization techniques combining genetic algorithms with deep learning and reinforcement learning approaches.

The session witnessed active participation from students, researchers, and faculty members. An interactive Q&A segment allowed participants to clarify their doubts and explore deeper insights into advanced AI methodologies. Students showed keen interest in understanding how they can implement such techniques in their academic projects and research work.

Overall, the expert talk was highly enriching and aligned with the university’s vision of fostering innovation and research excellence. It provided participants with both theoretical knowledge and practical perspectives, encouraging them to explore efficient AI solutions and contribute to the advancement of next-generation intelligent systems.

Department Name: School of Artificial Intelligence

Event Outcome 

  • Enhanced understanding of deep neural network optimization techniques, particularly model compression and acceleration. 
  • Participants gained practical insights into the application of genetic algorithms in AI model optimization. 
  • Improved awareness of deploying efficient AI models in real-world and resource-constrained environments such as IoT and edge devices. 
  • Encouraged students to explore research opportunities and innovative approaches in deep learning and evolutionary computing. 
  • Strengthened interaction between industry experts, faculty, and students, fostering a collaborative learning environment. 
  • Motivated participants to apply learned concepts in academic projects, internships, and research work.

Related Goal