Neuromorphic Circuit with Phase Change Synapse for Spike Timing Plasticity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current artificial neural network systems lack the ability to effectively mimic the dynamic synaptic plasticity and spike timing-dependent plasticity observed in biological neural networks, which are crucial for learning and behavior realization.
Innovation Solution
An artificial neuromorphic circuit comprising a synapse circuit with a phase change element, switches, and a post-neuron circuit that utilizes pulse signals, capacitors, and controllers to simulate synaptic conductance changes and generate firing signals, allowing for adaptive synaptic weight adjustments based on spike timing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional artificial neural network systems are used, then basic neural network functions are achieved, but the ability to mimic dynamic synaptic plasticity and spike timing-dependent plasticity is insufficient
Solution Approach 1:
The patent employs phase change elements that utilize phase transitions between crystalline and amorphous states to represent different synaptic weight values. This allows the circuit to dynamically adjust synaptic weights by controlling the phase state of the material, enabling spike timing-dependent plasticity without requiring complex control logic. The phase transition property directly encodes synaptic strength, simplifying the overall circuit architecture while achieving biological-like adaptability.
Solution Approach 2:
The invention changes the resistance parameter of the phase change element to represent synaptic weight. By applying voltage pulses that induce phase transitions, the resistance value changes discretely between high (amorphous) and low (crystalline) states, effectively mimicking synaptic weight adjustment. This parameter-based approach allows dynamic adaptability through simple voltage control rather than complex structural modifications.
2Adaptability or versatility
If synaptic weights are dynamically adjusted to mimic biological synapses, then learning capability is improved, but circuit complexity increases
Solution Approach 1:
The patent extracts the essential function of synaptic weight adjustment by isolating the phase change element as a dedicated component responsible for storing weight information. This separates the memory function (phase state) from the processing function (voltage pulse generation), allowing dynamic learning capability to be achieved through simple voltage control of the phase change element rather than complex interconnected circuits.
Solution Approach 2:
The invention copies the functional behavior of biological synapses by using the phase change element's resistance states to represent synaptic weights. The circuit replicates spike timing-dependent plasticity by comparing arrival times of voltage pulses and adjusting the phase change element's resistance accordingly, creating a simplified electronic copy of biological synaptic behavior without requiring complex circuitry.
3Use of energy by moving object
If phase change elements are used to represent synaptic weights, then energy efficiency is improved, but manufacturing precision requirements increase
Solution Approach 1:
The patent uses periodic voltage pulse signals to control the phase change element. By applying pulses at specific intervals and durations, the system can transition the phase change material between states with controlled energy input. This periodic puling approach is more energy-efficient than continuous voltage application, as energy is supplied only during state transitions rather than continuously maintaining the state.
Solution Approach 2:
The phase change element provides discrete, stable resistance states (high and low) corresponding to amorphous and crystalline phases. This binary-like state representation reduces the need for precise intermediate resistance control, thereby lowering manufacturing precision requirements while maintaining energy efficiency through stable state retention without continuous energy input.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables the creation of an artificial neural network system that learns and mimics biological neural networks by dynamically adjusting synaptic weights based on spike timing, effectively replicating the plasticity and causality of biological synapses.
Implementation Method 1
the phase change element has a first resistance value when the phase change element is in a first phase, and the phase change element has a second resistance value when the phase change element is in a second phase
Implementation Method 2
The capacitor is coupled to the switch circuit. The input terminal charges the capacitor through the switch circuit in response to the first pulse signal
Data Source
AI summary
Artificial neuromorphic circuit includes synapse circuit and post-neuron circuit. Synapse circuit includes phase change element, first switch, and second switch. First switch is coupled to phase change element, and is configured to receive first pulse signal. Second switch is coupled to phase change element. Input terminal of post-neuron circuit is coupled to switch circuit, and input terminal is coupled to phase change element. Input terminal charges capacitor through switch circuit in response to first pulse signal. Post-neuron circuit is configured to generate firing signal based on voltage level at input terminal and threshold voltage, and is further configured to generate first control signal and second control signal based on firing signal. Post-neuron circuit turns off switch circuit according to first control signal. Second control signal is configured to cooperate with second pulse signal to control second switch so as to control a state of phase change element.