← Back to Benchmarks
simmediumatarimetric · varies

Latent World Models For Intrinsically Motivated Exploration

Description

In this work we consider partially observable environments with sparse rewards. We present a self-supervised representation learning method for image-based observations, which arranges embeddings respecting temporal distance of observations. This representation is empirically robust to stochasticity and suitable for novelty detection from the error of a predictive forward model. We consider episodic and life-long uncertainties to guide the exploration. We propose to estimate the missing informat

Source

http://arxiv.org/abs/2010.02302v1