Variational Dynamic Mixtures

10/20/2020
by   Chen Qiu, et al.
0

Deep probabilistic time series forecasting models have become an integral part of machine learning. While several powerful generative models have been proposed, we provide evidence that their associated inference models are oftentimes too limited and cause the generative model to predict mode-averaged dynamics. Modeaveraging is problematic since many real-world sequences are highly multi-modal, and their averaged dynamics are unphysical (e.g., predicted taxi trajectories might run through buildings on the street map). To better capture multi-modality, we develop variational dynamic mixtures (VDM): a new variational family to infer sequential latent variables. The VDM approximate posterior at each time step is a mixture density network, whose parameters come from propagating multiple samples through a recurrent architecture. This results in an expressive multi-modal posterior approximation. In an empirical study, we show that VDM outperforms competing approaches on highly multi-modal datasets from different domains.

READ FULL TEXT

page 2

page 16

page 17

research
09/01/2023

Learning multi-modal generative models with permutation-invariant encoders and tighter variational bounds

Devising deep latent variable models for multi-modal data has been a lon...
research
06/07/2023

Multi-modal Latent Diffusion

Multi-modal data-sets are ubiquitous in modern applications, and multi-m...
research
08/24/2019

Conditional Flow Variational Autoencoders for Structured Sequence Prediction

Prediction of future states of the environment and interacting agents is...
research
12/01/2016

Piecewise Latent Variables for Neural Variational Text Processing

Advances in neural variational inference have facilitated the learning o...
research
11/27/2014

Learning Stochastic Recurrent Networks

Leveraging advances in variational inference, we propose to enhance recu...
research
03/11/2022

Dual reparametrized Variational Generative Model for Time-Series Forecasting

This paper propose DualVDT, a generative model for Time-series forecasti...
research
06/18/2018

Bayesian Prediction of Future Street Scenes through Importance Sampling based Optimization

For autonomous agents to successfully operate in the real world, anticip...

Please sign up or login with your details

Forgot password? Click here to reset