Motion Forecasting with Unlikelihood Training in Continuous Space

Deyao Zhu, Mohamed Zahran, Li Erran Li, Mohamed Elhoseiny

Research output: Contribution to conferencePaperpeer-review

7 Scopus citations


Motion forecasting is essential for making safe and intelligent decisions in robotic applications such as autonomous driving. Existing methods often formulate it as a sequence-to-sequence prediction problem, solved in an encoder-decoder framework with a maximum likelihood estimation objective. State-of-the-art models leverage contextual information, including the map and states of surrounding agents. However, we observe that they still assign a high probability to unlikely trajectories resulting in unsafe behaviors, including road boundary violations. Orthogonally, we propose a new objective, unlikelihood training, which forces predicted trajectories that conflict with contextual information to be assigned a lower probability. We demonstrate that our method can improve state-of-art models’ performance on the challenging nuScenes and Argoverse real-world trajectory forecasting datasets by avoiding up to 56% context-violated prediction and improving up to 9% prediction accuracy. Code is avaliable at

Original languageEnglish (US)
Number of pages10
StatePublished - 2021
Event5th Conference on Robot Learning, CoRL 2021 - London, United Kingdom
Duration: Nov 8 2021Nov 11 2021


Conference5th Conference on Robot Learning, CoRL 2021
Country/TerritoryUnited Kingdom

Bibliographical note

Funding Information:
This work is funded by a KAUST BAS/1/1685-01-0.

Publisher Copyright:
© 2021 Proceedings of Machine Learning Research. All rights reserved.


  • Autonomous Driving
  • Motion Forecasting

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability


Dive into the research topics of 'Motion Forecasting with Unlikelihood Training in Continuous Space'. Together they form a unique fingerprint.

Cite this