# Knowing When to Stop: Delay-Adaptive Spiking Neural Network Classifiers with Reliability Guarantees

arXiv: https://arxiv.org/abs/2305.11322 · HTML: https://mlsanity.com/paper/2305.11322/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 51 references that are in the map.
- First posted on arXiv 18 May 2023.

## Builds on (most cited first, 29 of 29)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification](https://mlsanity.com/paper/2107.07511/) (2021)
- [Distribution-Free Predictive Inference For Regression](https://mlsanity.com/paper/1604.04173/) (2016)
- [Surrogate Gradient Learning in Spiking Neural Networks](https://mlsanity.com/paper/1901.09948/) (2019)
- [BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks](https://mlsanity.com/paper/1709.01686/) (2017)
- [Conformal Prediction Under Covariate Shift](https://mlsanity.com/paper/1904.06019/) (2019)
- [Incorporating Learnable Membrane Time Constant to Enhance Learning of Spiking Neural Networks](https://mlsanity.com/paper/2007.05785/) (2020)
- [Conformal Risk Control](https://mlsanity.com/paper/2208.02814/) (2022)
- [Training Spiking Neural Networks Using Lessons From Deep Learning](https://mlsanity.com/paper/2109.12894/) (2021)
- [A Free Lunch From ANN: Towards Efficient, Accurate Spiking Neural Networks Calibration](https://mlsanity.com/paper/2106.06984/) (2021)
- [SPINN: Synergistic Progressive Inference of Neural Networks over Device and Cloud](https://mlsanity.com/paper/2008.06402/) (2020)
- [Learning Optimal Conformal Classifiers](https://mlsanity.com/paper/2110.09192/) (2021)
- [Conditional Deep Learning for Energy-Efficient and Enhanced Pattern Recognition](https://mlsanity.com/paper/1509.08971/) (2015)
- [Training Uncertainty-Aware Classifiers with Conformalized Deep Learning](https://mlsanity.com/paper/2205.05878/) (2022)
- [Few-shot Conformal Prediction with Auxiliary Tasks](https://mlsanity.com/paper/2102.08898/) (2021)
- [Exchangeability, Conformal Prediction, and Rank Tests](https://mlsanity.com/paper/2005.06095/) (2020)
- [SEENN: Towards Temporal Spiking Early-Exit Neural Networks](https://mlsanity.com/paper/2304.01230/) (2023)
- [An Introduction to Probabilistic Spiking Neural Networks: Probabilistic Models, Learning Rules, and Applications](https://mlsanity.com/paper/1910.01059/) (2019)
- [Conformal Prediction with Temporal Quantile Adjustments](https://mlsanity.com/paper/2205.09940/) (2022)
- [Few-Shot Calibration of Set Predictors via Meta-Learned Cross-Validation-Based Conformal Prediction](https://mlsanity.com/paper/2210.03067/) (2022)
- [Neuromorphic Wireless Cognition: Event-Driven Semantic Communications for Remote Inference](https://mlsanity.com/paper/2206.06047/) (2022)
- [Unleashing the Potential of Spiking Neural Networks by Dynamic Confidence](https://mlsanity.com/paper/2303.10276/) (2023)
- [Bayesian Continual Learning via Spiking Neural Networks](https://mlsanity.com/paper/2208.13723/) (2022)
- [Neuromorphic Integrated Sensing and Communications](https://mlsanity.com/paper/2209.11891/) (2022)
- [BiSNN: Training Spiking Neural Networks with Binary Weights via Bayesian Learning](https://mlsanity.com/paper/2012.08300/) (2020)
- [Spiking Neural Networks -- Part II: Detecting Spatio-Temporal Patterns](https://mlsanity.com/paper/2010.14217/) (2020)
- [Efficient Uncertainty Estimation in Spiking Neural Networks via MC-dropout](https://mlsanity.com/paper/2304.10191/) (2023)
- [Learning First-to-Spike Policies for Neuromorphic Control Using Policy Gradients](https://mlsanity.com/paper/1810.09977/) (2018)
- [Spiking Neural Networks -- Part III: Neuromorphic Communications](https://mlsanity.com/paper/2010.14220/) (2020)

## Built on by (most cited first, 8 of 8)

- [Federated Inference with Reliable Uncertainty Quantification over Wireless Channels via Conformal Prediction](https://mlsanity.com/paper/2308.04237/) (2023)
- [Uncertainty in Action: Confidence Elicitation in Embodied Agents](https://mlsanity.com/paper/2503.10628/) (2025)
- [Neuromorphic Wireless Split Computing with Multi-Level Spikes](https://mlsanity.com/paper/2411.04728/) (2024)
- [Direct Training Needs Regularisation: Anytime Optimal Inference Spiking Neural Network](https://mlsanity.com/paper/2405.00699/) (2024)
- [Think Short, Defer Smart, Act, and Repeat: Calibrated Reasoning and Uncertainty-Aware Deferral for Edge LLM Agents](https://mlsanity.com/paper/2607.26865/) (2026)
- [Stochastic Quantum Spiking Neural Networks with Quantum Memory and Local Learning](https://mlsanity.com/paper/2506.21324/) (2025)
- [Neuromorphic Wireless Split Computing with Resonate-and-Fire Neurons](https://mlsanity.com/paper/2506.20015/) (2025)
- [ELSA: An ELastic SNN Inference Architecture for Efficient Neuromorphic Computing](https://mlsanity.com/paper/2605.20802/) (2026)

## Abstract (arXiv)

Spiking neural networks (SNNs) process time-series data via internal event-driven neural dynamics. The energy consumption of an SNN depends on the number of spikes exchanged between neurons over the course of the input presentation. Typically, decisions are produced after the entire input sequence has been processed. This results in latency and energy consumption levels that are fairly uniform across inputs. However, as explored in recent work, SNNs can produce an early decision when the SNN model is sufficiently ``confident'', adapting delay and energy consumption to the difficulty of each example. Existing techniques are based on heuristic measures of confidence that do not provide reliability guarantees, potentially exiting too early. In this paper, we introduce a novel delay-adaptive SNN-based inference methodology that, wrapping around any pre-trained SNN classifier, provides guaranteed reliability for the decisions produced at input-dependent stopping times. The approach, dubbed SpikeCP, leverages tools from conformal prediction (CP). It entails minimal complexity increase as compared to the underlying SNN, requiring only additional thresholding and counting operations at run time. SpikeCP is also extended to integrate a CP-aware training phase that targets delay performance. Variants of CP based on alternative confidence correction schemes, from Bonferroni to Simes, are explored, and extensive experiments are described using the MNIST-DVS data set, DVS128 Gesture dataset, and CIFAR-10 dataset.