Q4 2021 Q&A Session — Part 2
In Part 2 of our Q&A, we continue answering questions from the call to action requested from our Telegram and Twitter community. In this edition, we take a look at Exchanges, DAO, Federated Learning and Oracles. Enjoy and as always, let us know if you have any questions!
Topic 1: New Exchanges
Q1. The team mentioned they are always looking for new liquidity / exchanges. Has there been any progress on this?
We are currently looking for new liquidity sources for PHB that may be a good compliment to Binance, including DEXs, local major exchanges, etc.
Q2. For exchanges that are not supporting the redom (redenomination), will we be delisted?
We are working with and in communication with all exchanges PHB is listed on to help cooperate with the redom, note that different exchanges may cooperate at different paces/speeds, and may be slower than Binance.
Topic 2: DAO (Decentralized Autonomous Organization)
Q3. When will the team provide a plan/transition to DAO structure? In what year will the project become a DAO
Currently a DAO is within plan, but is not necessary immediately — we will have Phoenix Global’s strategic interests as top priority and will make the transition (and thereby announce details) to a DAO when we feel like the timing and situation is optimal.
Q4. How does a DAO impact onboarding of Enterprises?
Enterprise and ecosystem cooperation and partnerships are one of the key factors we consider strategic, so it will be ensured that they will not be impacted if we initiate the transition to a DAO.
Topic 3: Federated Learning
Q5. Can the team provide more details on both the technical and current actual uses of FL going on with Phoenix and Pilots now. Practical examples will help further understanding, and assist with marketing this very important use case of the system.
Federated Learning simply put — is used when collaborative machine learning needs to take place without sacrificing data privacy. Instead of using the prevailing method (which is to aggregate the data first and centralize the training), we would train locally on local compute nodes and sync the results of the training back to the management node, where it “averages” the results and sends the new model to all the local nodes for retraining, and this is a continuous process.
How blockchain (and Phoenix Global) fits into the picture is:
- Process management —Ironically, there’s more centralized methods to FL, where one party manages the technology, processes, as well as “management node” for the participants of the FL (enterprises, organizations, or individuals). However there may not be a managing third party in all cases, in which case there needs to be a self executing protocol for the participants to use. This is where a perfect application of blockchain and namely smart contracts come into play. Training for AI models will always occur off-chain, however the process management can be facilitated on-chain.
- Governance — If FL can be can be managed in a completely decentralized manner, then there’s a few new problems and risks that can arise as a result, one of which is fraud and actors with malicious intent joining the FL process. For example, if a participant purposefully uses fake or problematic data in the process, we need ways to mitigate the risk and resolve the problem. This can be mitigated by 1) governance and voting mechanisms that serve to benefit participants as a whole 2) filtering mechanisms that can exclude outliers above a determined threshold.
Initially, FL in conjunction with blockchain will be a simple use case for Phoenix Global and its partners, however it’s quite effective in addressing the use cases above as well as a natural fit for blockchain technology. Currently it is not the consensus that FL needs to be used with blockchain but we, along with various other industry experts, academics, and organizations believe that it will be the trend and the preferred method of implementation in the future.
Q6. Will Phoenix Global be mentioned on the FLC website when it is launched? Will we be a prominent role in that program?
We are in the process of launching a full FLC website, and yes, Phoenix Global will be one of the keystone members. We’ll expect to launch the full website in November once the final stages of approval for FLC comes through from the Hong Kong regulators, which is when we are able to “officially” onboard members.
Q7. FL is new, and largely not understood by many. Plans for educational seminars — both technical and high level? Will there be documentation to assist with projects wanting to leverage FL on Phoenix and how to join/participate?
Webinars will be a part of FLC’s core activities along with research and collaborative projects.
Q8. When do you think FL will be more common place?
The niche technology and its use cases are growing fast, as well as global regulators increasingly tightening user privacy laws. We expect FL to rapidly proliferate in 2022.
Topic 4: Oracles
Q9. When a chain starts using the Phoenix Oracle what will Node stakers receive? Will this be updated in the whitepaper?
As we are still testing and developing the Phoenix Oracle, fees and incentive structures will need to be carefully planned out with network usage and primary use cases once reaches Beta stage. We will expect to release a new whitepaper/litepaper by then.
Q10. This is a very competitive space, with massive, established Oracles. What does the team plan to do gain share?
The oracle space still has significant room for market growth — large players such as Chainlink has its weaknesses and limitations in various aspects in terms of cross-chain, ease of use, etc. Us as a later entrant to the space, can benefit from understanding both their strengths and gaps and how they can be used to optimize our end product.
This concludes Part 2 of Q & A session, thank-you for submitting your questions and participating!
Join the discussion over in our Telegram, Reddit, and on Twitter.