Artificial Neuron Latency Rate Coding Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Artificial nervous systems face challenges in effectively discriminating between environmental stimuli due to limitations in existing methods that rely on either spike counting or latency coding, which are either robust against noise but slow or fast but less accurate.
Innovation Solution
Implementing a detector neuron that responds with a first spike based on latency coding and a second spike based on rate coding, using an interspike interval that functions as an error-checking signal to maintain accuracy and allow for error correction, enabling simultaneous use of both coding methods for enhanced stimulus discrimination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If spike counting is used for stimulus discrimination, then robustness against noise is improved, but response speed deteriorates
Solution Approach 1:
The patent segments the stimulus discrimination process into two distinct phases: a fast latency coding phase that provides immediate response, and a slower rate coding phase that provides noise-robust verification. This segmentation allows the system to achieve both speed and accuracy by performing operations in parallel at different time scales.
Solution Approach 2:
The patent implements preliminary action by using latency coding to make an initial stimulus discrimination decision quickly, then using rate coding to verify and correct this initial decision. The latency-based initial estimation allows the system to respond immediately while the subsequent rate-based verification ensures accuracy.
2Speed
If latency coding is used for stimulus discrimination, then response speed is improved, but measurement precision deteriorates
Solution Approach 1:
The patent implements feedback by using the rate-coded spike count to verify and correct the latency-coded stimulus discrimination result. The system compares the initial latency-based decision with the subsequent rate-based measurement, and uses this feedback to correct errors in the initial fast estimation.
Solution Approach 2:
The patent uses latency coding as a preliminary action to quickly estimate the stimulus type, then follows up with rate coding to verify and refine this estimation. This two-stage approach allows the system to achieve both rapid response and high precision.
3Device complexity
If single coding method is used, then device complexity is reduced, but adaptability deteriorates
Solution Approach 1:
The patent merges two different coding methods (latency coding and rate coding) into a unified stimulus discrimination system. By combining these complementary approaches, the system achieves both the speed of latency coding and the accuracy of rate coding, thereby improving adaptability without excessive complexity.
Solution Approach 2:
The patent creates a universal stimulus discrimination mechanism that can handle both fast response scenarios and high-precision requirements through its dual coding approach. The system adapts its operation mode based on the specific stimulus characteristics and timing requirements.
Data Source
AI summary
Methods and apparatus are provided for identifying environmental stimuli in an artificial nervous system using both spiking onset and spike counting. One example method of operating an artificial nervous system generally includes receiving a stimulus; generating, at an artificial neuron, a spike train of two or more spikes based at least in part on the stimulus; identifying the stimulus based at least in part on an onset of the spike train; and checking the identified stimulus based at least in part on a rate of the spikes in the spike train. In this manner, certain aspects of the present disclosure may respond with short response latencies and may also maintain accuracy by allowing for error correction.