Quantum

On linearity or non-linearity in machine learning for quantum chaotic dynamics

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 09/10/2026, 04:13:24 UTC.

Briefings

A 2026 arXiv preprint investigates whether machine learning can forecast quantum chaotic dynamics using a PXP chain benchmark. The study compares a nonlinear Transformer against a simple linear DLinear model. Results show the Transformer fails in scarred regimes while DLinear maintains high accuracy across all initial states, suggesting complex quantum evolution does not require complex forecasting models.

Key points

• The study uses a few-qubit PXP chain realized with Rydberg-atom arrays as a benchmark for quantum many-body dynamics.
• DLinear forecasting model maintains high accuracy across the entire family of initial states, including ergodic and scarred regimes.

Context and limitations

The research is limited to a few-qubit PXP chain benchmark; results may not generalize to larger systems or different quantum platforms.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 09/10/2026, 04:25:18 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeiho3vfdqlowfdivw5cez37pwflytab4fo7vhbsgmbww75tqyyieq4