Jon Tombs

About

Jon Tombs is a pioneering figure in the early development of neuromorphic engineering, with a focused research career dedicated to advancing on-chip learning for analogue VLSI neural networks. His most-cited work, "On-Chip Learning with Analogue VLSI Neural Networks" (1993), remains a foundational reference in the field, demonstrating through rigorous simulation how weight perturbation can be implemented as a viable learning scheme directly on analogue hardware. Tombs’ critical contribution was his realistic modeling of hardware limitations—particularly the constraints of synaptic weight precision—addressing the fundamental challenge of translating theoretical neural algorithms into practical, low-power, real-time silicon systems. This work, which has accumulated 10 citations, laid essential groundwork for subsequent generations of neuromorphic chips and edge-computing devices. By tackling the inherent trade-offs between computational fidelity and analogue circuit constraints, Tombs helped define the design principles that continue to guide researchers building efficient, brain-inspired computing hardware. His research remains a key reference for students and engineers exploring the intersection of machine learning, analogue circuit design, and VLSI systems.

Research Focus

Key Achievements

Top Papers

1. 1
   ON-CHIP LEARNING WITH ANALOGUE VLSI NEURAL NETWORKS

   10 citations · 1993

Key Collaborators

Contact & Links