Kartikay Garg

Atlanta Technical College, Georgia Institute of Technology

Papers

Total Citations

H-Index

About

Kartikay Garg’s research lies at the intersection of hardware-aware machine learning and bio-inspired computing, with a focus on enabling efficient, adaptive AI systems. His most impactful contribution is **GeneSys**, a framework introduced in his 2018 paper (22 citations) that brings continuous learning to hardware by evolving neural network topologies directly on-chip. This work addresses a critical bottleneck in modern deep learning—the reliance on static, hand-tuned architectures and massive labeled datasets—by allowing networks to adapt and improve over time without retraining from scratch. Garg’s earlier research on **swarm intelligence** (2013, 13 citations) provided a foundational overview of decentralized, self-organized AI systems, demonstrating his long-standing interest in adaptive, collective behaviors. Together, these works highlight his drive to move beyond conventional training paradigms toward more flexible, resource-efficient learning. GeneSys, in particular, has been recognized for its potential to bring lifelong learning to edge devices, a key step toward truly autonomous AI. Garg’s contributions are shaping how researchers think about neural architecture search and hardware-software co-design for continuous learning.