Neuromorphic Synapse Memelement Programming Logic

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional computing architectures are inefficient in terms of power consumption and space, and implementing sophisticated synaptic weight update rules for neuromorphic synapses is complex and costly, especially in nanoscale systems.

Innovation Solution

A neuromorphic synapse apparatus using a memelement with a specific programming characteristic and programming logic that generates signals for weight-dependent synaptic update efficacy, exploiting the internal physical state and history of the memelement to provide flexible and controlled synaptic dynamics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Power

If conventional computing architectures are used for neuromorphic computing, then processing power can be achieved, but power consumption and space requirements increase significantly

Engineering Contradiction:

Improvecomputing powerVSAvoidpower consumption

Core Design Contradiction:

PowerVSUse of energy by moving object

Solution Approach 1:

The patent replaces conventional von Neumann computing architecture with a neuromorphic architecture that uses memristive devices to directly implement synaptic weight storage and update functions. This substitution eliminates the separate memory and processing units, reducing data movement and associated power consumption while maintaining computational capability.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The memristive device serves multiple functions simultaneously: it stores synaptic weights, performs weight updates through voltage pulse application, and provides readout of weight values. This multi-functionality consolidates what would traditionally require separate components, reducing overall system power consumption and space requirements.

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Reliability

If sophisticated synaptic weight update rules are implemented in conventional systems, then accurate neuromorphic computation is achieved, but implementation complexity and cost increase

Engineering Contradiction:

Improvecomputational accuracyVSAvoidimplementation complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The memristive device automatically performs weight updates through its intrinsic physical properties. When voltage pulses are applied corresponding to pre- and post-synaptic neuron activity, the device's resistance changes accordingly through electroformation or other memristive mechanisms, eliminating the need for complex external control circuitry to implement STDP rules.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent exploits changes in the memristive device's resistance parameter in response to applied voltage pulses. By controlling the amplitude, duration, and timing of these pulses, sophisticated weight update rules are achieved through simple parameter modulation rather than complex logic circuits, reducing implementation complexity while maintaining computational accuracy.

Inventive Principle:

Principle #35Parameter changes

3Area of stationary object

If nanoscale memelement systems are used for synapse implementation, then space efficiency is improved, but manufacturing precision and control difficulty increase

Engineering Contradiction:

Improvesynapse areaVSAvoidprogramming precision

Core Design Contradiction:

Area of stationary objectVSManufacturing precision

Solution Approach 1:

The patent employs periodic voltage pulse trains for programming memristive synapses. By applying sequences of pulses with controlled amplitude and duration, cumulative resistance changes are achieved in a controlled manner, enabling precise weight programming despite nanoscale variability. This periodic action allows incremental adjustment of synaptic weights with high precision.

Inventive Principle:

Principle #19Periodic action

4Adaptability or versatility

If weight-dependent synaptic update efficacy is achieved through programming logic, then biological fidelity is improved, but device complexity increases

Engineering Contradiction:

Improvesynaptic plasticityVSAvoidprogramming logic complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements weight-dependent update efficacy through feedback mechanisms where the current resistance state of the memristive device influences the effect of subsequent programming pulses. The device's intrinsic properties create a feedback loop where previously programmed weights modulate future weight changes, achieving biologically faithful STDP and weight-dependent plasticity without requiring complex external control logic.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS9996793B2Neuromorphic synapses

Publication Date: 2018.06.12 SAMSUNG ELECTRONICS CO LTD

AI summary

Method to produce a neuromorphic synapse apparatus comprising a memelement for storing a synaptic weight, and programming logic. The memelement is adapted to exhibit a desired programming characteristic. The programming logic is responsive to a stimulus prompting update of the synaptic weight for generating a programming signal for programming the memelement to update said weight. The programming logic may be responsive to an input signal indicating an input weight-change value ΔWi, and may be adapted to generate a programming signal dependent on the input weight-change value ΔWi. The programming logic is adapted such that the programming signals exploit the programming characteristic of the memelement to provide a desired weight-dependent synaptic update efficacy.