A vision-only driving policy can be trained on-policy inside a learned world model simulator and deployed in a real ADAS, without hand-coded steering rules.
Impala: Scalable dis- tributed deep-rl with importance weighted actor-learner ar- chitectures
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Learning to Drive from a World Model
A vision-only driving policy can be trained on-policy inside a learned world model simulator and deployed in a real ADAS, without hand-coded steering rules.