PROJECT TITLE :
Visual Pattern Extraction Using Energy-Efficient “2-PCM Synapse” Neuromorphic Architecture
We tend to introduce a unique energy-efficient methodology “two-PCM Synapse” to use phase-amendment memory (PCM) as synapses in giant-scale neuromorphic systems. Our spiking neural network design exploits the gradual crystallization behavior of PCM devices for emulating both synaptic potentiation and synaptic depression. Unlike earlier makes an attempt to implement a biological-like spike-timing-dependent plasticity learning rule with PCM, we tend to use a simplified rule where long-term potentiation and long-term depression will each be made with one invariant crystallizing pulse. Our architecture is simulated on a special purpose event-based mostly simulator, employing a behavioral model for the PCM devices validated with electrical characterization. The system, comprising concerning two million synapses, directly learns from event-based dynamic vision sensors. When tested with real-life information, it's in a position to extract complicated and overlapping temporally correlated features like car trajectories on a freeway. Complete trajectories can be learned with a detection rate higher than ninety $p.c$. The synaptic programming power consumption of the system throughout learning is estimated and might be as low as one hundred nW for scaled down PCM technology. Robustness to device variability is also evidenced.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here