# Multimodal sequence dynamics and convergence optimization in dual-stream LSTM networks for complex physiological state estimation

## TL;DR

This paper introduces a new deep learning model that improves the accuracy of estimating complex physical states in volleyball training by better handling multimodal data.

## Contribution

The novel contribution is a Dual-Stream LSTM with temporal attention that reduces convergence issues and feature misalignment in multimodal sequence modeling.

## Key findings

- The model achieves 93.1% motion classification accuracy and 3.8% load modeling error.
- Velocity trajectory fitting reaches a coefficient of determination of 0.91 with peak deviation of 0.05 m/s.

## Abstract

The integration of virtual simulation with intelligent modeling is crucial for advancing the scientization and personalization of volleyball physical training. This study aims to overcome the convergence instability and feature misalignment in modeling multimodal kinematic and physiological sequences.

A dynamical framework based on a Dual-Stream Long Short-Term Memory network integrated with a temporal attention mechanism is proposed. The framework decouples heterogeneous feature learning and optimizes temporal weight distribution.

Experimental validation on complex motion state estimation demonstrates that the proposed model reduces load modeling error to 3.8% and achieves a motion classification accuracy of 93.1%. The velocity trajectory fitting coefficient of determination is 0.91 with a peak deviation of 0.05 m/s.

These results confirm the effectiveness of the attention-based DS-LSTM in optimizing multimodal sequence modeling for training state estimation and feedback.