Neural Core Circuit with Synaptic Interconnect for Neuromorphic Plasticity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current neuromorphic and synaptronic computation systems lack efficient methods to replicate the complex connectivity and plasticity of biological neurons, limiting their ability to effectively process and store experiences in a manner analogous to biological brains.

Innovation Solution

A neural core circuit with a synaptic interconnect network comprising electronic synapses, axon paths, and dendrite paths, along with a routing module, allows for configurable signal conduction between neurons, mimicking the functionality of biological synapses and enabling structural plasticity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If traditional digital models are used for computation, then ease of manufacture and operation are improved, but the ability to replicate biological neural functionality and structural plasticity deteriorates

Engineering Contradiction:

Improveease of manufactureVSAvoidability to replicate biological neural functionality

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The system segments neural computation into distinct electronic components: neurons that generate spikes, axons that transmit signals, synapses that modulate connectivity, and dendrites that receive inputs. This segmentation allows each component to be independently implemented using standard electronic circuits while collectively replicating biological neural functionality including structural plasticity through configurable synapse connections

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements a universal neuromorphic platform where a single electronic neural network architecture can perform multiple functions: pattern recognition, classification, and adaptive learning through structural plasticity. The configurable synapse connections allow the same hardware to adapt its connectivity structure dynamically, providing versatility while maintaining ease of manufacture through standardized circuit designs

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Adaptability or versatility

If complex connectivity and plasticity are replicated in biological-like systems, then adaptability and information processing capability are improved, but device complexity increases

Engineering Contradiction:

Improvestructural plasticityVSAvoiddevice complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent replaces complex biological mechanical systems with electronic circuit implementations. Synaptic connectivity and structural plasticity are achieved through electronic switching circuits and configurable connections rather than biological growth and differentiation. This substitution maintains adaptability while reducing device complexity through standardized electronic components and circuits

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system implements structural plasticity by dynamically changing connectivity parameters - specifically, the presence or absence of synaptic connections between neurons. This is achieved through configurable electronic switches that can be programmed to establish or remove connections based on activity patterns, providing biological-like adaptability through simple parameter changes rather than complex structural reorganization

Inventive Principle:

Principle #35Parameter changes

3Productivity

If biological neural networks are simulated with high fidelity, then information processing and storage capability are improved, but power consumption and hardware resource requirements increase

Engineering Contradiction:

Improveinformation processing capabilityVSAvoidpower consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The neuromorphic system uses periodic spiking activity rather than continuous analog signaling to transmit information between neurons. This event-driven periodic action reduces power consumption by keeping circuits in low-power states between spikes, while maintaining high information processing capability through efficient temporal coding and burst patterns that replicate biological neural activity

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system implements self-service through autonomous spike generation and propagation - neurons automatically generate action potentials when their membrane potential reaches threshold, and signals propagate through the network without external clocking or control. This self-organizing behavior reduces power consumption by eliminating the need for continuous external control signals while maintaining high productivity through efficient autonomous information processing

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS8868477B2Multi-compartment neurons with neural cores

Publication Date: 2014.10.21 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Embodiments of the invention provide a neural core circuit comprising a synaptic interconnect network including plural electronic synapses for interconnecting one or more source electronic neurons with one or more target electronic neurons. The interconnect network further includes multiple axon paths and multiple dendrite paths. Each synapse is at a cross-point junction of the interconnect network between a dendrite path and an axon path. The core circuit further comprises a routing module maintaining routing information. The routing module routes output from a source electronic neuron to one or more selected axon paths. Each synapse provides a configurable level of signal conduction from an axon path of a source electronic neuron to a dendrite path of a target electronic neuron.