Biomimetic olfactory processor with on-chip one-shot incremental learning for odor identification

Hong Chen, Ziyi Cheng, Dexuan Huo, Jilin Zhang, Kea‐Tiong TangResearch Square 2025 · DOI

TL;DR

Researchers have developed ANP-OB, a biomimetic olfactory processor that mimics the mammalian olfactory bulb to achieve one-shot incremental learning. This asynchronous, event-driven chip can identify new odors after a single exposure without forgetting previous ones, maintaining 99.8% accuracy under heavy noise while consuming less than 50 μW—making it the lowest-power olfactory processor reported to date.

What Problem Does This Paper Address?

Electronic noses (E-noses) are vital for public safety, food quality monitoring, and medical diagnostics. However, they face two massive hurdles: sensor drift and catastrophic forgetting.

Chemical sensors are notoriously unstable; their sensitivity and selectivity change over weeks or months (drift), causing traditional fixed-algorithm processors to lose accuracy over time. To fix this, most systems require retraining, but standard Artificial Neural Networks (ANNs) suffer from "catastrophic forgetting"—learning a new gas often wipes out the memory of old ones. Furthermore, the "dense" math required for deep learning is too power-hungry for the battery-operated, portable devices where E-noses are most needed.

Methods

The authors propose a Spiking Neural Network (SNN) architecture called ANP-OB (Asynchronous Neuromorphic Processor - Olfactory Bulb). Key innovations include:

  • Biomimetic Architecture: The chip mimics the two-layer structure of the mammalian olfactory bulb, using 128 "glomeruli." Each unit consists of an excitatory neuron (EN) coupled with five inhibitory neurons (INs).
  • Asynchronous "Click-based" Design: Unlike traditional chips that use a power-hungry global clock, ANP-OB is event-driven. It only consumes significant power when a "spike" (data) actually arrives.
  • Hybrid Learning Rule: It uses Spike-Timing-Dependent Plasticity (STDP) combined with lateral inhibition. This creates a "winner-take-all" effect where the first-responding neurons suppress others, polarizing synaptic weights to either their maximum or minimum values.
  • Power-Saving Hardware Tricks:
    • SSWU (Self-supervised Synapse Weight Update): Dynamically skips redundant weight updates if the change is negligible, reducing training power by 71.6%.
  • SWCS (Synapse Weight Compression Storage): Exploits the fact that weights polarize to the extremes, mapping 8-bit weights to a 4-bit representation to save 50% of storage space without losing accuracy.
  • LUT-based Leakage: Replaces complex real-time leaky-integrate-and-fire (LIF) calculations with a simple Look-Up Table (LUT).

Datasets

The team validated the processor using three major publicly available datasets:

  1. UCI Gas Sensor Array Drift Dataset: A 6-class dataset spanning 36 months to test long-term drift compensation.
  2. UCI Open Sampling Settings Dataset: A 10-class dataset used to test one-shot learning under varying noise levels.
  3. CSIRO (FOX) 20-Chemical Dataset: A complex 20-class dataset to test the limits of the processor's discrimination ability.

Key Insights and Ideas

  • Timing is Everything: By using a rate-based spike precedence coding scheme, the processor turns gas concentration into the timing of spikes. This allows the biological principle of lateral inhibition to naturally "filter" noise and identify the most prominent chemical signatures.
  • Polarization as a Feature: In many AI models, weight precision is a burden. Here, the authors noticed that their biomimetic learning caused 95% of weights to hit their ceiling or floor. They turned this "limitation" into a compression feature (SWCS), proving that biological-style sparsity is a hardware engineer's best friend.
  • On-Chip vs. Off-Chip: The paper demonstrates that on-chip learning is not just a luxury—it is a necessity for E-noses. By learning "on the fly," the chip improved accuracy by 24.8% compared to static off-chip models when dealing with aged, drifting sensors.

Key Results

  • Accuracy: Achieved 99.8% accuracy in a 10-class task under 60% noise. Even with "one-shot" learning (learning a gas from a single sample), it maintained 93.2% accuracy under 70% noise.
  • Power Efficiency: Inference power consumption is <50 μW. The asynchronous LIF neuron design alone provided a 92.9% reduction in dynamic power compared to traditional synchronous designs.
  • Drift Robustness: On real-world 3-year drift data, the system maintained over 98% accuracy by incrementally learning from new samples, effectively "solving" the sensor drift problem.
  • Speed: The chip requires only 0.67 ms to identify a gas sample, significantly faster than GPU-based simulations.

Limitations and Future Work

While a breakthrough in efficiency, the authors acknowledge several constraints:

  • Algorithmic Rigidity: The current hardware is hard-wired for the STDP algorithm. While it excels at gas identification, it performs poorly on unrelated tasks like gesture recognition.
  • Chemical Similarity: The "winner-take-all" mechanism can struggle to distinguish between chemicals with nearly identical compositions, as they trigger nearly identical spike patterns.
  • Scalability: The current implementation is at a "modest scale" (768 neurons). Future work will likely involve scaling the architecture to handle higher-dimensional data and more complex chemical mixtures.

Summary updated

Abstract

<title>Abstract</title> Electronic noses (E-noses) mimic the mammalian olfactory system and have been used to detect gases in public safety, disease diagnosis, food safety and so on. However, existing E-noses lack of adaptive learning ability, making them susceptible to sensor drift caused by environmental changes, leading to a progressive decline in recognition accuracy over time. Moreover, unlike mammalian olfactory systems with robust learning abilities and adaptation, current E-noses cannot continuously and rapidly learn new odors profiles or compensate for environmental perturbations. To address the limitations, we propose a novel biomimetic olfactory processor ANP-OB which achieves rapid learning capabilities by mimicking the mammalian olfactory bulb. The ANP-OB is able to remember a new gas after learning it only once, and does not forget previously learned gases. With one-shot incremental learning, ANP-OB achieves a recognition accuracy of 99.8% in a 10-class gas recognition task under 60% noise. Furthermore, its power consumption is less than 50 μW with event-driven asynchronous circuits, making it the lowest-power olfactory processor to our knowledge.
BibTeX
@inproceedings{chen2025biomimetic,
  author = {Hong Chen and Ziyi Cheng and Dexuan Huo and Jilin Zhang and Kea‐Tiong Tang},
  title = {{Biomimetic olfactory processor with on-chip one-shot incremental learning for odor identification}},
  booktitle = {Research Square},
  year = {2025},
  doi = {10.21203/rs.3.rs-7646030/v1},
}
Figure