Home/ CSE/ Spiking Neural Networks (SNNs)
CSE · Seminar 07 · Brain-inspired, event-driven computation

Spiking Neural Networks (SNNs)

SNNs communicate with discrete spikes over time rather than continuous activations, enabling ultra-low-power, event-driven inference on neuromorphic hardware.

SNNneuromorphicLIF neuronspike codingSTDP

Conventional deep networks pass continuous real numbers between layers. Spiking neural networks instead mimic biological neurons: information is carried by the timing of discrete, binary spikes. Because a neuron only does work when it spikes, computation is event-driven and extremely energy-efficient — the basis of neuromorphic computing.

Working principle

The workhorse model is the Leaky Integrate-and-Fire (LIF) neuron. Its membrane potential integrates incoming weighted spikes and leaks toward rest over time. When the potential crosses a threshold the neuron emits a spike and resets. Information is encoded as spike rate or precise timing; learning can use biologically-inspired STDP or surrogate-gradient backpropagation through time.

yesfire→resetInput spikes Σ wᵢ·sᵢ(t)Membrane V(t) integrate + leakV ≥ Vth?Emit spikeReset VLeaky Integrate-and-Fire neuron dynamics
Figure 1. LIF neuron. The membrane potential accumulates input and leaks; crossing the threshold produces a spike and a reset, giving temporal, event-driven behaviour.
Table 1. Artificial NN vs. Spiking NN
PropertyANNSNN
SignalContinuous valuesBinary spikes over time
TimeStatic forward passTemporal dynamics
EnergyDense MAC operationsEvent-driven, sparse
HardwareGPU/TPUNeuromorphic (Loihi, etc.)
TrainingBackpropSTDP / surrogate gradients
Trade-offOn neuromorphic chips, SNNs can cut inference energy by one to two orders of magnitude for sparse, event-based workloads — but training accuracy still lags ANNs, the field's main open problem.

Applications

  • Always-on keyword spotting and gesture recognition on tiny power budgets
  • Event-camera (DVS) vision for high-speed, low-latency robotics
  • Edge sensor fusion where milliwatts matter

References & further reading

  1. Maass, “Networks of spiking neurons: the third generation of neural network models,” Neural Networks, 1997.
  2. Davies et al., “Loihi: A Neuromorphic Manycore Processor with On-Chip Learning,” IEEE Micro, 2018.
  3. Neftci et al., “Surrogate Gradient Learning in Spiking Neural Networks,” IEEE SPM, 2019.