Accurate and Diverse Sampling of Sequences Based on a "Best of Many" Sample Objective (2018-06-01T00:00:00.000000Z)

TL;DR

This work addresses challenges in a Gaussian Latent Variable model for sequence prediction with a "Best of Many" sample objective that leads to more accurate and more diverse predictions that better capture the true variations in real-world sequence data.

Abstract

For autonomous agents to successfully operate in the real world, anticipation of future events and states of their environment is a key competence. This problem has been formalized as a sequence extrapolation problem, where a number of observations are used to predict the sequence into the future. Real-world scenarios demand a model of uncertainty of such predictions, as predictions become increasingly uncertain - in particular on long time horizons. While impressive results have been shown on point estimates, scenarios that induce multi-modal distributions over future sequences remain challenging. Our work addresses these challenges in a Gaussian Latent Variable model for sequence prediction. Our core contribution is a "Best of Many" sample objective that leads to more accurate and more diverse predictions that better capture the true variations in real-world sequence data. Beyond our analysis of improved model fit, our models also empirically outperform prior work on three diverse tasks ranging from traffic scenes to weather data.

Authors

B. Schiele

35 papers

Mario Fritz

14 papers

Apratim Bhattacharyya

3 papers

Accurate and Diverse Sampling of Sequences Based on a "Best of Many" Sample Objective

TL;DR

Abstract

Authors

References28 items

Deep Learning for Precipitation Nowcasting: A Benchmark and A New Model

A Generative Model of People in Clothing

Motion Prediction Under Multimodality with Conditional Stochastic Networks

DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents

Simplified Stochastic Feedforward Neural Networks

End-to-End Learning of Driving Models from Large-Scale Video Datasets

Learning Social Etiquette: Human Trajectory Understanding In Crowded Scenes

Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks

Social LSTM: Human Trajectory Prediction in Crowded Spaces

Stochastic Multiple Choice Learning for Training Diverse Deep Ensembles

Unsupervised Learning for Physical Interaction through Video Prediction

Variational Inference for Monte Carlo Objectives

Learning Structured Output Representation using Deep Conditional Generative Models

Predicting Multiple Structured Visual Interpretations

MuProp: Unbiased Backpropagation for Stochastic Neural Networks

Importance Weighted Autoencoders

Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting

Sequence to Sequence Learning with Neural Networks

Semi-supervised Learning with Deep Generative Models

Techniques for Learning Binary Stochastic Feedforward Neural Networks

Stochastic Backpropagation and Approximate Inference in Deep Generative Models

Learning Stochastic Feedforward Neural Networks

Generating Sequences With Recurrent Neural Networks

Multiple Choice Learning: Learning to Produce Multiple Structured Outputs

Calculus: The Elements

AUTO-ENCODING VARIATIONAL BAYES

The mnist sequence dataset

Unsupervised modeling of the movement of basketball players using a Deep Generative Model

Field of Study

Journal Information

Name

Page