# Traversing Latent Space using Decision Ferns

## TL;DR

This paper introduces a novel framework for controlled exploration of latent spaces using a Variational Autoencoder and decision ferns, enabling smooth traversal and applications in spatial transformations and kinematics prediction.

## Contribution

It presents a new controller module for latent space traversal and an end-to-end trainable system that enhances manipulation of latent representations.

## Key findings

- Effective latent space traversal demonstrated
- Application to spatial transformations shown
- Predictive kinematics in video sequences achieved

## Abstract

The practice of transforming raw data to a feature space so that inference can be performed in that space has been popular for many years. Recently, rapid progress in deep neural networks has given both researchers and practitioners enhanced methods that increase the richness of feature representations, be it from images, text or speech. In this work we show how a constructed latent space can be explored in a controlled manner and argue that this complements well founded inference methods. For constructing the latent space a Variational Autoencoder is used. We present a novel controller module that allows for smooth traversal in the latent space and construct an end-to-end trainable framework. We explore the applicability of our method for performing spatial transformations as well as kinematics for predicting future latent vectors of a video sequence.