Search papers, labs, and topics across Lattice.
This paper introduces PRISM, a novel policy representation for robot motor control that explicitly incorporates polynomial interactions among observable physical variables, enhancing the learning of action-relevant cues. By integrating a factorized polynomial module into standard MLP architectures, PRISM effectively captures higher-order interactions without the computational burden of enumerating all polynomial terms. The approach demonstrates significant performance improvements in humanoid locomotion and contact-rich manipulation tasks compared to traditional MLP policies, indicating that simply increasing model capacity is insufficient for capturing interaction structures in motor control.
Polynomial representations in robot control can outperform larger MLPs, revealing that interaction structures are crucial for effective motor policies.
Robot policies are typically MLPs mapping observations to actions. Yet robot observations are physical variables, and many action-relevant cues arise not from individual variables but from their interactions; power, inertial effects, contact, slip, and compliance depend on products among observable signals. We introduce PRISM, a policy representation that makes polynomial interactions among observable physical variables explicit, learnable, and compact. Rather than listing all polynomial terms, PRISM uses a factorized polynomial module to expose higher-order interaction features efficiently. In reinforcement learning, it keeps the standard MLP backbone but applies a gradually activated element-wise polynomial function after it. In imitation learning, it replaces linear proprioceptive conditioning in Diffusion Policy with a polynomial layer trained end-to-end. Across humanoid locomotion and contact-rich manipulation, PRISM improves performance over standard MLP policies and larger MLPs with matched capacity, showing that interaction structure cannot be replaced by capacity alone. It also yields sensorless compliant behavior without force, wrench, tactile input, contact labels, or admittance control. These results suggest that polynomial representations should become a standard architectural choice for embodied motor control. The project page is available at https://lsh3163.github.io/prism/
