An instance-centric representation with local frames, relative positional encodings, and adaptive reward transformation in adversarial IRL yields scalable, accurate, and robust behavior models for multi-agent driving simulation.
We evaluate two variants: 1) trained with c= 5 , as proposed in [6], and 2) trained with our proposed adaptive reward offset, defined in (3)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Toward Efficient and Robust Behavior Models for Multi-Agent Driving Simulation
An instance-centric representation with local frames, relative positional encodings, and adaptive reward transformation in adversarial IRL yields scalable, accurate, and robust behavior models for multi-agent driving simulation.