FEDERATED LEARNING DEMO · READ-ONLY WALKTHROUGH

Updates move. Records stay local.

Train across sites that can never pool their records. Each round ships only the update, ends in a signed proof, and leaves auditors a bundle they can examine. When EU AI Act provenance enforcement begins in August 2026, you can show who trained on what, and under whose approval.

APPROVED PARTIES
18

Active collaborators visible on the federation dashboard before the next round starts.

COMPLETED + LIVE ROUNDS
32

Round history stays visible so operators can compare drift, throughput, and review state.

BUDGET LEFT
0.84

Remaining privacy budget carried forward from the live privacy-budget editor.

THE LOOP

From approved roster to signed bundle.

Five steps. Every one carries who joined, who approved, and the proof of what aggregated.

STEP 01

Register a node and capture the provider, region, telemetry, and join token.

STEP 02

Approve the participant list before starting a new round.

STEP 03

Collect local model updates while source records stay inside each organization boundary.

STEP 04

Verify the aggregation proof and privacy-budget spend-down on one record.

STEP 05

Generate the audit export and hand the signed bundle to reviewers.

PARTICIPANT MATRIX

Local metrics and privacy checks stay visible.

Every site contributes an update; every site keeps its records.

SiteHealthMetricBudgetContributionFailed check
North Clinichealthy0.910.1118%none
West Hospitalhealthy0.880.0914%none
Finance EUreview0.830.1612%latency
APAC Regiondropped0.710.220%budget
AGGREGATION PROGRESS

Five steps. One signed round.

STEP 01

Approve participant roster

● DONE
STEP 02

Collect encrypted local updates

● DONE
STEP 03

Run secure aggregation

● DONE
STEP 04

Verify privacy checks

○ REVIEW
STEP 05

Sign round packet

○ REVIEW
ROUND 32 · FEDAVG + SECURE AGGREGATION
AGGREGATION IN FLIGHT
APPROVED
18
SUBMITTED
17
PROOF HASH
agg_9a3c1f2b7d44
ROUND READING · SEED DATA · LIVE IN A PILOT
SEEDED SURFACE PREVIEWS

Three previews. One seeded state.

The figures here are illustrative seed data. In a pilot, every count reads from your own rounds.

DASHBOARD SURFACE

Dashboard preview

  • 18 approved parties are visible before the next round starts.
  • 32 total rounds are already tracked in the shared history.
  • Budget remaining is carried forward from the platform privacy-budget editor (0.84).
ROUND DETAIL SURFACE

Round detail preview

  • Round 32 is currently aggregation in flight.
  • 17 updates are in, 2 parties are flagged for follow-up.
  • Proof hash agg_9a3c1f2b7d44 stays visible beside latency and communication cost.
PRIVACY-BUDGET SURFACE

Privacy-budget preview

  • Current policy snapshot: epsilon 1.4, delta 1e-5.
  • 63% of the tracked budget is already consumed across seeded rounds.
  • Privacy-budget spend-down is reviewed before the next round is approved.
SEED-BACKED SCENARIOS

Where records can never leave.

HEALTHCARE

Hospital network coordination

Multiple clinical sites contribute local imaging updates while protected health information stays inside each hospital boundary.

Coordinators see round health, approvals, and proof on one release record.

FINANCE

Fraud signal sharing

Regional risk teams improve a shared model without centralizing transaction histories into a single warehouse.

Security and compliance reviewers can inspect proof before the next learning round starts.

CROSS-BORDER

Regional model operations

Country or partner-specific nodes keep local policy boundaries intact while contributing to a shared objective.

Privacy budget spend-down stays understandable for non-cryptographers.

NEXT PATH

Federation, with proof attached.

Read-only here. To run these rounds on your own sites and keep the weights they produce, start a pilot.

Federated Learning | Privacy-preserving coordination | AuraOne