Validators

Validators secure DeepNode’s execution layer by verifying miner outputs, enforcing model correctness, and ensuring that all tasks in the network meet the required quality thresholds.

The flow below outlines how validators participate in the incentive loop:

---

1. Role of Validators

See Validators for a more in depth description.

---

2. Incentive Flow

The validator incentive system works through a structured pipeline:

Step 1: Tasks Enter the Domain

Consumers generate model request via the platform or API.

Step 2: Miners Process Tasks

Miners perform the compute work and submit their results. These results are forwarded to validators for auditing.

Step 3: Validators Run Verification

Validators check whether miners outputs are correct.

• If validation is correct: → The task is approved → Rewards are emitted to all relevant actors (miners, validators, creators/backers)

• If validation fails: → Miner trust score is reduced → Validators notify the domain → The task may be reassigned or penalized depending on severity

Step 4: Reward Emission

When validators approve a task, the reward engine triggers:

• Miners receive their utility-weighted reward

• Validators receive their share of emission

• Stakers receive a share proportional to their stake

• The domain receives its revenue allocation

• Model owners/backers receive their configured share

Step 5: Staking and Delegation

Validators are strengthened by stakers who delegate their $DN to validator operators. Delegators earn a percentage of validator rewards, and validators accumulate trust and influence proportional to delegated stake.

This ensures:

• long-term alignment

• economic security

• a robust and decentralized validator set

---

3. Fee Routing

As shown in the diagram:

• A portion of revenue (y%) flows into Fees & Payment Processors

• A portion of fees may route to Variable Burn (d%) depending on network governance

For a complete overview of fees in the network see: Fees.

---

4. Why Validators Matter

Validators:

• ensure correctness

• maintain economic sustainability

• prevent manipulation

• preserve model integrity

• protect users from malicious outputs

• participate in governance

They are the backbone of the entire $DN incentive mechanism.