AIP-2: Availability-Weighted Reward Mechanism
Adjust validator rewards proportionally to their availability score to incentivize consistent uptime and reliable participation.
| Authors | Javad Rajabzadeh (@ja7ad) |
|---|---|
| Discussion | View Discussion AIP-2 |
| Category | Core |
| Created | 2025-09-13 |
Table of Contents
Abstract
This proposal introduces a mechanism to scale validator rewards based on their availability score. Validators with higher availability scores will earn a larger portion of their base reward, while those with lower scores will receive proportionally reduced rewards. This adjustment aims to incentivize validators to maintain consistent uptime and reliable participation, thereby improving network performance and security.
Motivation
Currently, validator rewards are distributed evenly regardless of their reputation or availability score, as long as the validator is active in the committee. This can result in validators with suboptimal availability still receiving the same reward as highly reliable validators. By linking rewards to availability, we align economic incentives with the network’s need for dependable participation, discourage semi-active validators, and create a fairer reward system based on actual contribution.
Specification
- Availability Score:
- A value between $0.0$ and $1.0$ representing the validator’s recent performance and uptime.
- Define $\tau = \tfrac{2}{3} \approx 0.666667$.
- Scores $< \tau$ → temporary exclusion from the committee (no rewards).
- Scores $\ge \tau$ → eligible for scaled rewards.
- Reward Scaling Formula:
- Let $base\_reward$ be the validator’s share of the block reward (e.g., if the protocol allocates 70% of each block reward to validators and 30% to the foundation, then $base\_reward$ = 0.7 AUM).
-
The adjusted reward is calculated as:
\[reward = base\\_reward × (α + (1 − α) × score)\] - Where:
- $α$ = fixed proportion (e.g., 0.5) ensuring a minimum reward floor.
- $score$ = availability score of the validator (0.0 ≤ score ≤ 1.0).
- Examples (with $base\_reward = 0.7$, $α = 0.5$):
- $score = 1.0$ → $reward = 0.7\,\mathrm{AUM}$
- $score = 0.98$ → $reward = 0.693\,\mathrm{AUM}$
- $score = 0.666667$ ($\approx \tau$) → $reward \approx 0.5833\,\mathrm{AUM}$
- $score = 0.6$ (below $\tau$) → excluded → $reward = 0$
- $score = 0.0$ (below $\tau$) → excluded → $reward = 0$
This approach ensures validators with excellent availability maximize their rewards, while poorly performing validators earn less; validators below the threshold receive no rewards while temporarily excluded.
Backwards Compatibility
No backward compatibility issues found. Existing reward distribution rules remain compatible, with only the distribution formula adjusted.
Test Cases
- Validator $score = 1.0$, $reward = 0.7$
- Validator $score = 0.98$, $reward ≈ 0.693$
- Validator $score = 0.6$ (below $\tau$), $reward = 0$
- Validator $score = 0.0$ (below $\tau$), $reward = 0$
- Validator $score < 0.666667$ and excluded from committee → $reward = 0$
Security Considerations
- Sybil Resistance: Validators may attempt to reset identity to bypass low scores. This must be mitigated by slashing, stake requirements, or bonding periods.
- Score Manipulation: Availability scoring must be robust against manipulation and accurately reflect validator performance.
- Fairness: Ensures consistent uptime is rewarded but prevents harsh penalties that could discourage participation.
- Network Resilience: By rewarding reliability, the system promotes a stronger validator set and reduces the risk of downtime.
Copyright
Copyright and related rights waived via CC0.