By identifying low-dimensional manifolds and coupling patterns in robot co-design landscapes, the authors develop an adaptive algorithm that infers task-specific structure during search, achieving 36% better performance and over 100x sample efficiency than benchmarks.
DiffTaichi: Differentiable Programming for Physical Simulation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Identifying Inductive Biases for Robot Co-Design
By identifying low-dimensional manifolds and coupling patterns in robot co-design landscapes, the authors develop an adaptive algorithm that infers task-specific structure during search, achieving 36% better performance and over 100x sample efficiency than benchmarks.