pith:HEW4KX4R
Learning to Maximize Quantum Neural Network Expressivity via Effective Rank
Quantum neural networks achieve full expressivity when their circuits, data inputs, and measurements are all optimized to the theoretical maximum.
arxiv:2506.15375 v4 · 2025-06-18 · quant-ph · physics.comp-ph
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{HEW4KX4RRIRZXH4APC6MWUDRSB}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
κ can saturate its theoretical upper bound, d_n=4^n-1, for an n-qubit system when each of the three factors (circuit architecture, input data distributions, and measurement protocols) is optimally expressive.
That the count of effectively independent parameters (κ) directly corresponds to the functional expressivity that determines performance on variational tasks, rather than being a proxy that may miss other limiting factors such as trainability or noise.
Introduces effective rank κ to quantify QNN expressivity and applies reinforcement learning with a transformer agent to optimize circuit architectures for higher κ.
Receipt and verification
| First computed | 2026-05-28T01:04:28.731758Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
392dc55f918a239b9f8078bccb5071904f6ed570c54ef3732af0eba81cd73607
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/HEW4KX4RRIRZXH4APC6MWUDRSB \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 392dc55f918a239b9f8078bccb5071904f6ed570c54ef3732af0eba81cd73607
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "997a1e8637061037cbf2c14fcf698c48aad66f6ba3156165e4609f3b3bc74a3a",
"cross_cats_sorted": [
"physics.comp-ph"
],
"license": "http://creativecommons.org/publicdomain/zero/1.0/",
"primary_cat": "quant-ph",
"submitted_at": "2025-06-18T11:45:57Z",
"title_canon_sha256": "f748a3087944e4c4c0609f28706b0c3e7aedef75e562c7afe86648ac2a024af1"
},
"schema_version": "1.0",
"source": {
"id": "2506.15375",
"kind": "arxiv",
"version": 4
}
}