NeuroMemFPP: A recurrent neural approach for memory-aware parameter estimation in fractional Poisson process
By: Neha Gupta, Aditya Maheshwari
In this paper, we propose a recurrent neural network (RNN)-based framework for estimating the parameters of the fractional Poisson process (FPP), which models event arrivals with memory and long-range dependence. The Long Short-Term Memory (LSTM) network estimates the key parameters $μ>0$ and $β\in(0,1)$ from sequences of inter-arrival times, effectively modeling their temporal dependencies. Our experiments on synthetic data show that the proposed approach reduces the mean squared error (MSE) by about 55.3\% compared to the traditional method of moments (MOM) and performs reliably across different training conditions. We tested the method on two real-world high-frequency datasets: emergency call records from Montgomery County, PA, and AAPL stock trading data. The results show that the LSTM can effectively track daily patterns and parameter changes, indicating its effectiveness on real-world data with complex time dependencies.
Similar Papers
CNN-Based Automated Parameter Extraction Framework for Modeling Memristive Devices
Emerging Technologies
Makes computer memory faster and easier to design.
Accelerated Execution of Bayesian Neural Networks using a Single Probabilistic Forward Pass and Code Generation
Machine Learning (CS)
Makes AI safer by knowing when it's wrong.
Predicting and forecasting reactivity and flux using long short-term memory models in pebble bed reactors during run-in
Systems and Control
Predicts how nuclear reactors will work safely.