# Modeling continuous-time stochastic processes using $\mathcal{N}$-Curve mixtures

## TL;DR

This paper introduces a neural network model called $
$-Curve mixtures for continuous-time stochastic processes that can generate smooth, multi-modal sequence predictions efficiently, improving over existing methods in trajectory and motion modeling.

## Contribution

The paper proposes a novel $
$-Curve mixture model based on Mixture Density Networks with Bézier curves, enabling multi-modal, smooth predictions in a single inference step.

## Key findings

- Outperforms state-of-the-art models in human trajectory prediction
- Effective in human motion modeling tasks
- Reduces reliance on Monte Carlo simulations for multi-step predictions

## Abstract

Representations of sequential data are commonly based on the assumption that observed sequences are realizations of an unknown underlying stochastic process, where the learning problem includes determination of the model parameters. In this context the model must be able to capture the multi-modal nature of the data, without blurring between modes. This property is essential for applications like trajectory prediction or human motion modeling. Towards this end, a neural network model for continuous-time stochastic processes usable for sequence prediction is proposed. The model is based on Mixture Density Networks using B\'ezier curves with Gaussian random variables as control points (abbrev.: $\mathcal{N}$-Curves). Key advantages of the model include the ability of generating smooth multi-mode predictions in a single inference step which reduces the need for Monte Carlo simulation, as required in many multi-step prediction models, based on state-of-the-art neural networks. Essential properties of the proposed approach are illustrated by several toy examples and the task of multi-step sequence prediction. Further, the model performance is evaluated on two real world use-cases, i.e. human trajectory prediction and human motion modeling, outperforming different state-of-the-art models.