AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

ShadowNet for Data-Centric Quantum System Learning

Yuxuan Du, Yibo Yang, Tongliang Liu, Zhouchen Lin, David A. Clifton, Bernard Ghanem, Dacheng Tao

TPAMI 2026 · Probabilistic and scientific modeling › Dynamical systems · approach: Physics-informed dynamics learning

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
increase training set size from 8 to 200 (ablation)attention ShadowNet · TFIM QSTtest ground energy estimation error: 0.94 -> 0.02Improved
increase training set size from 8 to 200 (ablation)attention ShadowNet · TFIM QSTtest fidelity: 0.81 -> 1Improved
increase training set size from 100 to 800 (M=500) (ablation)attention ShadowNet · TFIM+XXZ QSTtest fidelity: 0.878 -> 0.995Improved
increase training set size from 100 to 800 (M=500) (ablation)attention ShadowNet · TFIM+XXZ QSTground energy estimation error: 1.361 -> 0.044Improved
increase shot number from 500 to 2000 at n=100 (ablation)attention ShadowNet · TFIM+XXZ QSTtest fidelity: 0.878 -> 0.933Improved
increase shot number from 500 to 2000 at n=800 (ablation)attention ShadowNet · TFIM+XXZ QSTtest fidelity: 0.995 -> 0.996No clear effect
use ShadowNet instead of classical shadows (n=800, M=500)ShadowNet · TFIM+XXZ QSTground energy estimation error: 0.477 -> 0.044Improved
use ShadowNet instead of classical shadows (n=800, M=2000)ShadowNet · TFIM+XXZ QSTground energy estimation error: 0.204 -> 0.038Improved
use attention instead of convolution (n=100) (ablation)ShadowNet · TFIM+XXZ QSTtrain fidelity: 0.706 -> near 1Improved
use attention instead of convolution (n=800) (ablation)ShadowNet · TFIM+XXZ QSTtrain fidelity: 0.914 -> near 1Improved
add classical shadows to noise-parameter features (ablation)attention ShadowNet · noisy GHZ DFE (N=60)test loss: 0.0147 -> 0.00013Improved
add noise parameters to classical shadow features (ablation)attention ShadowNet · noisy GHZ DFE (N=60)test loss: 0.0196 -> 0.00013Improved

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