Maintaining Price Stability with a Decentralized Price Oracle

The Double N Protocol leverages a decentralized price oracle and Supply Decision Algorithm (SDA) to dynamically adjust the token supply and maintain price stability. The oracle monitors the secondary market to assess supply and demand, triggering actions to expand or contract the supply as needed.

Case A: Expanding the Supply

1. The SDA identifies the required additional supply to stabilize the price.

2. Stakeholders vote on the appropriate adjustment using their staked Double N.

3. The system calculates the median of the votes and distributes the additional Double N proportionally to the stakers who participated.

4. Votes significantly outside the expected range may result in reduced rewards to ensure system integrity.

The newly issued tokens alleviate upward price pressure, restoring balance to the system.

Case B: Contracting the Supply

1. The SDA determines the amount of supply reduction needed to stabilize the price.

2. Stakeholders vote on the required adjustment.

3. The system issues a 14-day bond, offering guaranteed interest rates.

4. Bonds are auctioned, and stakeholders can participate proportionally to their stake.

5. Unsold bonds are reissued with increased interest rates.

6. The treasury buys back and burns Double N tokens until price stability is achieved.

Participants in the bond issuance are rewarded with both their initial investment and accrued interest, aligning incentives with the protocol’s stability goals.

Supply Decision Algorithm (SDA)

The SDA is a pivotal feature of the Double N ecosystem, designed to dynamically regulate supply based on market conditions. This algorithm employs advanced AI-driven methodologies to detect market imbalances and implement corrective measures.

Time Series Data Analysis Engine (Deep learning Engine). This engine analyses and processes time series data then computes real time decision through its historical data and decisions. The input for this engine consists of [Price, Volume, PVC, etc.], and the output follows the form of [Tax_rate, Exchange_rate]. With the “Gradient Descent Algorithm”, loss function is computed as Loss = -exp(Profit) with the function of profit as Profit = ( delta P * Decision ) / Volume. The algorithm described above is operated with the AdamOptimizer, and the LSTM Cell is used to analyse and to train time series data.Comment

Classifier, Computation Engine (Probabilistic Engine) Classifier uses the method of the Gaussian Process, GMM, K-MC and the Regression to divide the situation of the coin market and the blockchain using historical data. Through this classified decision, numerical data including information of tax rate and exchange rate is arithmetically operated by computation engine. (3) Reinforcement Learning Engine (RL Engine) The learning of the SDA algorithm is based on the ‘Q-learning’ algorithm based on the value of the ‘Q-function’ of the ‘Deep-Q-Network (DQN)’ that belongs to the ‘Agent’. ‘Agent’ selects the action by the ‘Epsilon greedy method’. This action is applied to the ‘Zipett blockchain environment’ to calculate the ‘Reward (USDZ minting)’ and the ‘Next state’. This history is stored in the ‘Replay memory’ and updates the internal variables of DQN through the ‘Experience replay’ method. In the above learning and internal variable updating process, ‘Q-learning’ is performed through the optimization method by the ‘slope descent method’ as follows. The diagram below is the structure of the SDA engine and the error function used in the learning.

Key Functions of the SDA

• Market Detection: Continuously analyzes liquidity, transaction volume, and price volatility to assess market conditions.

• Dynamic Supply Adjustment: Utilizes machine learning to forecast and implement optimal supply changes.

• Consensus-Based Implementation: Decisions are validated via Proof of Stake mechanisms, ensuring transparency and trust.

• Market Execution: Adjustments are reflected in the market, and participants are rewarded proportionally to their contributions.

---

Incentives for Double N Stakeholders

Stakeholders play a vital role in ensuring the stability and security of the Double N ecosystem and are rewarded through multiple mechanisms:

• Transaction Fees: A portion of transaction fees collected on the network is distributed to stakers.

• Seigniorage Rewards: During supply expansions, newly minted Double N is distributed among active stakeholders.

• Revenue Sharing: Income generated from the Double N-A ecosystem and related platforms is allocated to stakeholders as dividends.