arXiv Open Access 2017

Learning A Physical Long-term Predictor

Sebastien Ehrhardt Aron Monszpart Niloy J. Mitra Andrea Vedaldi
Lihat Sumber

Abstrak

Evolution has resulted in highly developed abilities in many natural intelligences to quickly and accurately predict mechanical phenomena. Humans have successfully developed laws of physics to abstract and model such mechanical phenomena. In the context of artificial intelligence, a recent line of work has focused on estimating physical parameters based on sensory data and use them in physical simulators to make long-term predictions. In contrast, we investigate the effectiveness of a single neural network for end-to-end long-term prediction of mechanical phenomena. Based on extensive evaluation, we demonstrate that such networks can outperform alternate approaches having even access to ground-truth physical simulators, especially when some physical parameters are unobserved or not known a-priori. Further, our network outputs a distribution of outcomes to capture the inherent uncertainty in the data. Our approach demonstrates for the first time the possibility of making actionable long-term predictions from sensor data without requiring to explicitly model the underlying physical laws.

Topik & Kata Kunci

Penulis (4)

S

Sebastien Ehrhardt

A

Aron Monszpart

N

Niloy J. Mitra

A

Andrea Vedaldi

Format Sitasi

Ehrhardt, S., Monszpart, A., Mitra, N.J., Vedaldi, A. (2017). Learning A Physical Long-term Predictor. https://arxiv.org/abs/1703.00247

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2017
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓