Related Experiment Video
Updated: Jan 16, 2026
10:45
Time-dependent Increase in the Network Response to the Stimulation of Neuronal Cell Cultures on Micro-electrode Arrays
Published on: May 29, 2017
10.3K
On training networks of monostable multivibrator timer neurons
Lars Keuninckx1, Matthias Hartmann1, Paul Detterer2
1Interuniversity Microelectronics Centre (IMEC), Kapeldreef 75, Leuven, 3001, Belgium.
Summary
Neuromorphic hardware can overcome synaptic addition bottlenecks using simple timers (monostable multivibrators) that OR inputs. This enables efficient, event-driven networks for classification tasks with low energy consumption.
Area of Science:
- Neuromorphic Engineering
- Digital Hardware Design
- Artificial Neural Networks
Background:
- Current neuromorphic hardware faces limitations due to synaptic addition, restricting parallelization and processing speed.
- This bottleneck hinders the development of highly efficient and scalable artificial intelligence systems.
Purpose of the Study:
- To introduce a novel neuromorphic network architecture that mitigates the synaptic addition bottleneck.
- To demonstrate the effectiveness of this new approach in solving complex classification tasks.
Main Methods:
- Implementation of a network using monostable multivibrator timers, where synaptic inputs are OR-ed, bypassing traditional addition.
- Development of a training algorithm for fully binarized, event-driven recurrent networks of these timers.
- Resolution of temporally overlapping input events within the training algorithm.
Main Results:
- Successful training of the proposed network for classification tasks, including MNIST handwritten digits, Google Soli radar gestures, IBM DVS128 gestures, and Yin-Yang.
- Achieved high test accuracy (98.61% on MNIST) with a reconfigurable network of 500 units.
- Estimated low energy consumption of 855pJ per inference for the MNIST task (excluding readout layer) on a 28nm process.
Conclusions:
- Monostable multivibrator networks offer a viable solution to the synaptic addition bottleneck in neuromorphic hardware.
- This approach enables efficient, event-driven processing suitable for various classification challenges.
- The demonstrated low energy consumption highlights the potential for power-efficient neuromorphic computing.
Related Concept Videos
01:25Neural Circuits
2.6K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
2.6K
01:25Network Function of a Circuit
642
Frequency response analysis in electrical circuits provides vital insights into a circuit's behavior as the frequency of the input signal changes. The transfer function, a mathematical tool, is instrumental in understanding this behavior. It defines the relationship between phasor output and input and comes in four types: voltage gain, current gain, transfer impedance, and transfer admittance. The critical components of the transfer function are the poles and zeros.
642
01:22Applications of RC Circuits
4.0K
A relaxation oscillator is one of the applications of RC circuits. A neon lamp relaxation oscillator comprises a capacitor, a resistor, a voltage source, and a lamp. The lamp acts like an open circuit, with infinite resistance until the potential difference across the lamp reaches a specific voltage. At that voltage, the lamp acts like a short circuit with zero resistance, and the capacitor discharges through the lamp, thus producing light. Once the capacitor is fully discharged through the...
4.0K
01:20Motor Unit Stimulation
3.5K
When the neuron of a motor unit fires an action potential, it triggers a series of events, leading to a twitch contraction in the muscle fibers. The process of excitation-contraction coupling is crucial in relaying the action potential to the muscle fibers.
The latent period of contraction marks the onset of excitation-contraction coupling, when the action potential propagates across the sarcolemma, preparing the muscle fibers for contraction. As the fibers enter the contraction phase, the...
The latent period of contraction marks the onset of excitation-contraction coupling, when the action potential propagates across the sarcolemma, preparing the muscle fibers for contraction. As the fibers enter the contraction phase, the...
3.5K
01:35Long-term Potentiation
58.3K
Long-term potentiation, or LTP, is one of the ways by which synaptic plasticity—changes in the strength of chemical synapses—can occur in the brain. LTP is the process of synaptic strengthening that occurs over time between pre- and postsynaptic neuronal connections. The synaptic strengthening of LTP works in opposition to the synaptic weakening of long-term depression (LTD) and together are the main mechanisms that underlie learning and memory.
58.3K
01:28Electrical Synapses
10.2K
Electrical synapses found in all nervous systems play important and unique roles. In these synapses, the presynaptic and postsynaptic membranes are very close together (3.5 nm) and are actually physically connected by channel proteins forming gap junctions.
Gap junctions allow the current to pass directly from one cell to the next. In contrast, in the chemical synapse, the neurotransmitters carry the information through the synaptic cleft from one neuron to the next. They consist of two...
Gap junctions allow the current to pass directly from one cell to the next. In contrast, in the chemical synapse, the neurotransmitters carry the information through the synaptic cleft from one neuron to the next. They consist of two...
10.2K