AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

CINOC: Cardinality-Invariant Neural Operator Policies for Scalable PDE Control

Pietro Zanotta, Dibakar Roy Sarkar, Honghui Zheng, Somdatta Goswami, Jan Drgona

ICML 2026 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Imitation and skill learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add differentiable-physics training with a neural operator policy (CINOC) instead of best RLCINOC · KS 2Denergy: 0.50 ± 0.24 -> 0.37 ± 0.19Improved
replace DPC with CINOCCINOC · KS 2Denergy: 9.74 ± 2.32 -> 0.37 ± 0.19Improved
use CINOC instead of DPC in obstacle-constrained Heat 2DCINOC · Heat 2D (Obs)MSE: 2.7e-4 -> 1.2e-4Improved
use CINOC instead of best RL (MAPPO) on Heat 2D (Obs)CINOC · Heat 2D (Obs)MSE: 0.004 -> 1.2e-4Improved
use CINOC instead of DPC on Heat 1DCINOC · Heat 1DMSE: 7.9e-5 -> 2.9e-4Hurt
use CINOC instead of RL baselines on Density ControlCINOC · Density ControlMML: 0.002 vs 0.002 (DPC)No clear effect
use CINOC instead of PPO on Density ControlCINOC · Density ControlMML: 0.005 -> 0.002Improved
apply control to KS 2D versus uncontrolled evolutionCINOC · KS 2Denergy: 12.16 -> 0.20Improved
use CINOC instead of MAPPO on KS 1DCINOC · KS 1Denergy: 0.13 ± 0.13 vs 0.13 ± 0.12No clear effect
use CINOC instead of DPC on Turbulence 2DCINOC · Turbulence 2Denstrophy: 3.28 ± 2.19 -> 3.23 ± 2.22Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy