MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611110334 A) filed by Chandigarh University on September 14, 2026, for Event-Driven Gradient-Free Synaptic Plasticity For Sparse Neuromorphic Process.
Inventors include Shaman Sharma; and Deepak Kumar.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: Event-Driven Gradient-Free Synaptic Plasticity for Sparse Neuromorphic Processing relates to a neuromorphic processing system using discrete spike events for computation and learning. A plurality of spiking neuron units integrates input events and generates output spikes, while local plasticity logic (103, 104) evaluates relative pre- and post-synaptic spike timing and updates synaptic states according to a spike-timing-dependent rule without gradient-based backpropagation. A temporal-contrastive evaluation favours earlier desired output spikes relative to undesired output spikes and may favour reduced output spike activity. Structural adaptation (105) prunes underutilized connections and establishes new pathways in response to repeated temporal spike patterns. The system operates asynchronously without a global training clock and can be implemented using neuromorphic electronic devices, including memristive synaptic elements, for sparse event-driven processing in embedded and edge applications.
Disclaimer: Curated by HT Syndication.
