Sumit Bam Shreshta
Papers
1
Total Citations
5
H-Index
1
About
Sumit Bam Shrestha is a leading researcher at the intersection of neuromorphic computing and efficient deep learning, with a focus on developing biologically inspired architectures for next-generation hardware. His most notable contribution is the introduction of a Diagonal Structured State Space Model (S5) optimized for Intel’s Loihi 2 neuromorphic processor, a breakthrough that enables highly efficient streaming sequence processing. This work, published in 2025 and already garnering 5 citations, addresses the unsustainable energy costs of conventional deep learning accelerators like GPUs by demonstrating how novel architectures can achieve competitive performance with dramatically lower power consumption. Shrestha’s research bridges the gap between algorithmic innovation and hardware implementation, tackling the fundamental challenge of adapting deep learning models to non-von Neumann computing paradigms. By showing that state space models can be effectively deployed on event-driven neuromorphic chips, he has opened new pathways for real-time, low-power applications in edge computing and robotics. His work is particularly impactful for students and researchers exploring energy-efficient AI, as it provides a concrete example of how algorithmic design must co-evolve with emerging hardware to sustain the growth of intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1