Encryptability As a Coordinate Choice: Depth-One Homomorphic Federated Learning of Quantum Neural Networks

Marcel Mordarski; Nathan Mani; Arshad Patel; William Knottenbelt; Roberto Bondesan

Encryptability As a Coordinate Choice: Depth-One Homomorphic Federated Learning of Quantum Neural Networks

Marcel Mordarski, Nathan Mani, Arshad Patel, William Knottenbelt, Roberto Bondesan

arXivarxiv/2609.30581Sep 24, 2026quant-phscore 0

Encrypted training relies on keeping server-side updates low-degree. This constraint traditionally excludes models whose weights inhabit a compact Lie group (notably variational quantum circuits, where every trainable weight is an $\mathrm{SU(2)}$ rotation). Expressed in Euler angles or discrete alphabets, these updates appear transcendental, historically demanding prohibitive costs: one client--server round per gate, or upwards of $25{,}000$ operations per weight. This penalty is strictly an artefact of coordinates. In the unit-quaternion (spin) chart, group composition is exactly bilinear…

Source JSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2609.30581
curl -X POST /api/v1/records/arxiv/2609.30581/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2609.30581/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'