Paper: arXiv 2512.05868
Authors: Brian Ezinwoke, Oliver Rhodes
Abstract
Modern high-frequency trading (HFT) environments are characterized by sudden price spikes that present both risk and opportunity, but conventional financial models often fail to capture the required fine temporal structure. Spiking Neural Networks (SNNs) offer a biologically inspired framework well-suited to these challenges due to their natural ability to process discrete events and preserve millisecond-scale timing. This work investigates the application of SNNs to high-frequency price-spike forecasting, enhancing performance via robust hyperparameter tuning with Bayesian Optimization (BO). This work converts high-frequency stock data into spike trains and evaluates three architectures: an established unsupervised STDP-trained SNN, a novel SNN with explicit inhibitory competition, and a supervised backpropagation network. BO was driven by a novel objective, Penalized Spike Accuracy (PSA), designed to ensure a network’s predicted price spike rate aligns with the empirical rate of price events. Simulated trading demonstrated that models optimized with PSA consistently outperformed their Spike Accuracy (SA)-tuned counterparts and baselines. Specifically, the extended SNN model with PSA achieved the highest cumulative return (76.8%) in simple backtesting, significantly surpassing the supervised alternative (42.54% return). These results validate the potential of spiking networks, when robustly tuned with task-specific objectives, for effective price spike forecasting in HFT.
Complexity vs Empirical Score
- Math Complexity: 7.5/10
- Empirical Rigor: 6.0/10 (capped from 6.5: transaction costs not modelled; see penalty flags)
- Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 6.5)
Why this score: The paper employs advanced mathematics including STDP equations and Bayesian optimization formulas, placing it in the high math category. Empirically, it demonstrates robust backtesting with cumulative returns, specific datasets, and hyperparameter tuning metrics, qualifying as high rigor.
Research Flowchart
flowchart TD
A["Research Goal:<br>Predict HFT Price Spikes with SNNs"] --> B["Data Processing:<br>Convert High-Frequency Stock Data into Spike Trains"]
B --> C{"Methodology:<br>Model Architecture & Optimization"}
C --> D["Unsupervised STDP SNN"]
C --> E["Extended SNN w/ Inhibitory Competition"]
C --> F["Supervised Backpropagation Network"]
D & E & F --> G["Optimization Engine:<br>Bayesian Optimization driven by Penalized Spike Accuracy PSA"]
G --> H["Key Findings/Outcomes:<br>PSA-optimized models outperform baselines.<br>Extended SNN + PSA achieved 76.8% cumulative return vs. 42.54% for supervised alternative."]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper applies spiking neural networks to high‑frequency price‑spike forecasting and improves performance via Bayesian‑optimized hyperparameter tuning, introducing a Penalized Spike Accuracy objective.
Source sentence
This work investigates the application of SNNs to high-frequency price-spike forecasting, enhancing performance via robust hyperparameter tuning with Bayesian Optimization (BO).
Assumes
No explicit assumptions found in the text.
Data
2015-02 – 2015-03 · high-frequency stock data · equities · intraday
- High-frequency AAPL stock price and volume data market-prices unknown
Source sentence
High-frequency $AAPL stock price and volume data with microsecond precision from the 19 trading days of February 2015 was used for training and evaluating the model.
Costs
Not modeled
Finds 2
Sharpe 19.71 · total return 76.8% · max drawdown 2.7%
- The extended SNN model tuned with the Penalized Spike Accuracy metric achieved the highest cumulative return of 76.8% in simple backtesting, far exceeding the supervised alternative's 42.5% return. performance
Source sentence
Specifically, the extended SNN model with PSA achieved the highest cumulative return (76.8%) in simple backtesting, significantly surpassing the supervised alternative (42.5% return).
- Models optimized with the PSA objective consistently outperformed their Spike Accuracy‑tuned counterparts and baseline models. performance
Source sentence
Simulated trading demonstrated that models optimized with PSA consistently outperformed their Spike Accuracy (SA)-tuned counterparts and baselines.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 12 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (62)
- High-Frequency Trading and Financial Time-Series Prediction with Spiking Neural Networks (2023)
- Spiking neural networks for financial data prediction (2013) — link
- Hebbian Learning in Networks of Spiking Neurons Using Temporal Coding (2006)
- Methodology based on spiking neural networks for univariate time-series forecasting (2024) — link
- Algorithms for hyper-parameter optimization (2011)
- STDP for clustering spatio-temporal data (2018)
- Spiking Neural Networks and Their Applications: A Review (2022) — link
- Error-Backpropagation in Temporally Encode Networks of Spiking Neurons (2000)
and 54 more
- Training Spiking Neural Networks Using Lessons From Deep Learning (2023) — link
- Spiking Neural Networks and online learning: An overview and perspectives (2020) — link
- Towards spike-based machine intelligence with neuromorphic computing (2019) — link
- An Approach to Time Series Forecasting With Spiking Neural Networks (2023)
- Hyperopt: Distributed asynchronous hyperparameter optimization in Python (2013)
- Neuroflow: A general purpose spiking neural network simulation platform using customizable processors (2016) — link
- Networks of spiking neurons: The third generation of neural network models (1997) — link
- SWAT: An Unsupervised SNN Training Algorithm for Classification Problems (2008)
- Financial Time Series Forecasting Using Spiking Neural Networks (2020)
- Receptive Fields, Binocular Interaction and Functional Architecture in the Cat’s Visual Cortex (1962) — link
- Poisson Model of Spike Generation (2000)
- An Unsupervised STDP-based Spiking Neural Network Inspired by Biologically Plausible Learning Rules and Connections (2023) — link
- The Spike-Timing Dependence of Plasticity (2012) — link
- A Review of Learning in Biologically Plausible Spiking Neural Networks (2020) — link
- Unsupervised learning of digit recognition using spike-timing-dependent plasticity (2015) — link
- Deep learning in spiking neural networks (2018) — link
- Making a Science of Model Search: Hyperparameter Optimization in Hundreds of Dimensions for Vision Architectures (2013)
- Optuna: A Next-Generation Hyperparameter Optimization Framework (2019)
- Artificial neural networks for non-stationary time series (2004)
- Financial time series forecasting with deep learning: A systematic literature review: 2005–2019 (2020) — link
- Financial Time-series Analysis: a Brief Overview (2007) — link
- Financial Time Series Forecasting: Comparison of Traditional and Spiking Neural Networks (2021)
- Characteristics and Forecast of High-Frequency Trading (2021)
- Improving the Prediction Accuracy of Financial Time Series by Using Multineural Network Systems and Enhanced Data Preprocessing (1998)
- Efficient and Effective Time-Series Forecasting with Spiking Neural Networks (2024)
- DIET-SNN: Direct Input Encoding With Leakage and Threshold Optimization in Deep Spiking Neural Networks (2020)
- High-Frequency Trading Using Machine Learning: A Comprehensive Analysis (2024)
- High-Frequency Trading (HFT) (2023)
- Data Characteristics for High-Frequency Trading Systems (2015) — link
- Comparison of Artificial and Spiking Neural Networks on Digital Hardware (2021) — link
- Spiking Neural Networks for Nonlinear Regression (2022) — link
- SSTDP: Supervised Spike Timing Dependent Plasticity for Efficient Spiking Neural Network Training (2021) — link
- STDP-based Unsupervised Feature Learning using Convolution-over-time in Spiking Neural Networks for Energy-Efficient Neuromorphic Computing (2018) — link
- Spiking Back Propagation Multilayer Neural Network Design for Predicting Unpredictable Stock Market Prices with Time Series Analysis (2010)
- Financial Time Series Forecasting using CNN and Transformer — → 2304.04912
- Toward Large-scale Spiking Neural Networks: A Comprehensive Survey and Future Directions (2024) — link
- Supervised learning of spatial features with STDP and homeostasis using Spiking Neural Networks on SpiNNaker (2024) — link
- A Tutorial on Bayesian Optimization (2018)
- Introduction to Time Series and Forecasting (2016)
- A Carbon Price Forecasting Model Based on Variational Mode Decomposition and Spiking Neural Networks (2016)
- Statsmodels: Econometric and Statistical Modeling with Python (2010)
- Time Series Analysis by State Space Methods (2012)
- Reconsidering the energy efficiency of spiking neural networks (2024)
- Practical Bayesian Optimization of Machine Learning Algorithms (2012)
- Bayesian-based Hyperparameter Optimization for Spiking Neuromorphic Systems (2019) — link
- The Nutter’s guide to \LaTeX (1994)
- URL s in BibTeX bibliographies (2012)
- Finance basics for tough times (2009)
- Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs (1997) — link
- Synaptic modification by correlated activity: Hebb’s postulate revisited (1998)
- A neuronal learning rule for sub-millisecond temporal coding (1996) — link
- The Organization of Behavior: A Neuropsychological Theory (1949)
- Competitive Hebbian learning through spike-timing-dependent synaptic plasticity (2000) — link
- Adam: A Method for Stochastic Optimization (2017) — link
1 of 62 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.