Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - biophil

Pages: [1] 2 3 4 5 6 7 8 ... 59
1
On the one side 201907-uccs-research-project does create a problem, where there is actually is no problem, but doesn't solve any of our real problems.
The BitShares community is already aware MCR = 1.5 and MSSR = 1.01 is the sweet spot.

A very quick answer: how do you know there is no problem? The community chose the older chain parameters and then crypto winter caused most BitAssets to GS. Since then, you've adjusted the parameters to shift more risk onto BitAsset holders. Why would those new parameters be better?

A question back to you: How do you know that 1.5 and 1.01 are the sweet spot? Why not 1.2 and 1?

2
@bench: you spoke very negatively about this worker in other threads, which leads me to believe that your mind is made up already. But I'm happy to answer anyway.

@biophil:
1. How can your theoretical model be applied to a market with external parameters?

There are two broad ways: the first is called a Monte Carlo simulation, and we're laying the groundwork for this already. We simulate the model with many different inputs for the various parameters, and see what outcomes occur. For example, if we run Monte Carlo simulations for many different market conditions on the current chain parameters and most of the outcomes have good pegging and liquidity, then that validates the current chain parameters.

The second way is more difficult, but we will try given funding: there is a major area of control theory called "robust control" where they study whether a system performs well for all kinds of disturbances. It is possible to show that certain kinds of control systems (airplane autopilots, for example) are robust and can avoid spiraling out of control no matter what nature does to them. We may be able to apply this theory to the BitShares system and describe exactly what kinds of market conditions are "safe" for BitAsset price pegs.

Quote
2. How does your worker solve systematic market problems?

Our view is this: BitAssets will never be widely adopted if their users don't have confidence. If someone comes to you and says "why is MSSR set to 1.01?" and your answer is "we picked it randomly and it's been working fine for a while," that doesn't instill confidence. If instead you can say "we have run extensive simulations and advanced analysis and we have found that 1.01 provides the best balance between keeping the price down in bull markets and keeping the price up in bear markets," then you'll have something.

Quote
3. What options/solutions do you want to research?

I've been clear about this from the start, and I'm still proposing the same thing now that I was when I first announced. Take a look at the document linked in the first post of this thread.

3
I am Dr. Philip N. Brown, professor of computer science at the University of Colorado and lead on the 201907-uccs-research-project worker which aims to fund research on BitAssets and create simulation models to help improve price pegging and collateralization for BitAssets.

Current Project Status and activities

Before any funds came to the worker proposal, I wrote a paper to send to the Decentralized 2019 conference; the paper has already been made available to the BitShares community here: http://cs.uccs.edu/~pbrown2/papers/Decentralized_Preprint.pdf.

Work is progressing well. The PhD student who is attached to the project has been learning about financial modeling and this week we'll start developing our BitAsset simulator. The simulator will be a testbed which we will use to look at the effects of different choices of chain parameters, and as time goes on we will expand it to look at more advanced stabilization techniques also.

If the project receives sufficient funding (approximately $40k, to ensure the researcher/student is fully funded for at least one year), we will make sure that the simulation code is available for the BitShares community to use to test various aspects of pegging behavior on their own.

Accounting Information

This worker proposal is a BitShares Blockchain Foundation escrow worker, which means that any community member is free to examine the accounting information at any time: https://www.bitshares.foundation/workers/2019-07-uccs-research-project

We have not yet requested a payment from the foundation, since so far only $14,000 has been allocated and it would be considerably better for us to receive a larger amount as our first disbursement.

Project costs

The proposal originally requested $150,000 to fully fund a 2-year project. Approximately half of that money was slated to be used to fund the student's efforts; this means the original proposal asked for about $80,000 to pay for a full-time PhD student/researcher to work on the project for 2 years. The bulk of the remainder of the funds were to be used to pay me to devote some of my summers to the project full-time.

I will continue to stress that $150,000 would set this up as a very successful project, but I also understand that since the BTS price has decreased so much in the past months, the community is now hesitant to part with this much money all at once.

In light of this, I want the cn-vote community to know that this project will give the community results even if it is only funded at a level of $40,000. Even this extremely low level of funding would ensure that the project's student would be fully funded and completely committed to the project for a full year.

Thus, I am asking the cn-vote community to consider voting the research worker back in until at least $40,000 have been raised. At current prices, this should take about 3 weeks, and it would be a good balance between funding the project and reducing BTS inflation. $40,000 would allow us to make significant progress (build a complete simulator for the community to use for experiments, and then analyze the simulation results deeply and draw some conclusions about how to set the chain parameters optimally), and then if the BTS price increases in the future we can introduce a new worker to continue the work.

Some Questions

Q1. If $40,000 is enough for the project to generate value, why does the original proposal ask for $150,000?
A: At universities, students do most of the research work; at my university, a full-time student costs at least $40,000 per year. The extra money was going to be used to "buy my time" from the University so that I could devote large blocks of my time to the project as well. So: $40,000 buys the project a student for a year; $150,000 buys the project a student for 2 years plus a professor (me) full-time for a total of four months. If project costs are cut, I will always sacrifice my time first and keep the student working. However, of course the more effort that is devoted to the project, the more successful it will be.

Q2. Why does this worker ask for all the money at the start of the project?
A: I wish there were a better way to do this, but unfortunately the way research funding works at universities is that the money must be committed at the start of the project. Part of the reason for this is to make sure that the student is confident in their ability to continue working throughout the project. However, the BitShares community needs to understand that my reputation as a professor is at stake if I mis-manage the money: a very important part of my career is that when I tell people I'm going to be able to get research done, I absolutely need to deliver on my promise.

Q3. Why is this worker important for the BitShares community?
A: A (relatively) short answer is this: this worker brings 3 core benefits to the BitShares community.
First, we will provide specific recommendations to the community about how chain parameters such as MCR and MSSR should be chosen, along with detailed simulation results and mathematical analysis to support our recommendations. In turn, when the BitShares community advertises their product, they will be in a very strong position to say "we've had this looked at, and here are all the reasons why we do things the way we do them." This should make BitShares a stronger competitor.
Second, we will publish and present our findings at well-respected academic conferences (and tell everybody that BitShares sponsored us), which will directly increase the global awareness of the BitShares system.
Third, by funding high-level research, BitShares will help to define what kinds of problems are "interesting" and worth studying; when other researchers see money going to a particular topic, it increases their interest in that topic. Of course this benefit is the least tangible, but it really is an important piece of funding university research that many people forget about.

As always, please direct your questions to me and I will do my best to answer them.

4
I am pro research (0.5 year), but why should we finance a student and travel costs?

The worker already has 15k $ and it is unclear, which benefits the community gets from the worker.

One solution is here with the settlement fund: https://github.com/bitshares/bsips/issues/182
(Time to respond to my BSIP)
 
The main problem BTS has, is the deficit spending by overpriced workers, which dump BTS and don't bring any real value to the ecosystem.

Bench, maybe your mind is made up and an explanation wouldn't change it, but here you go:

In the US university system, practically all research is done by advanced PhD students; those students are directed by professors and paid a small stipend by external funding sources. The research is the student.

I wrote the Decentralized 2019 paper on my own time because I think the BitAsset problem is a very interesting one and I was making a good-faith effort to launch the project before funding arrived. Unfortunately, my own time is very limited, and unless an external funding source "buys" some of my official University time, I can't devote the many many hours that it will take to do justice to this problem. This is where the student comes in: he will spend the bulk of the time on the project, and he must be the first one to get paid -- I won't take a cent for myself from this worker unless he is funded full-time. Half-time funding isn't really appropriate for a PhD student.

To put this another way: your BSIP deserves more than my quick comments, but my in-depth comments will take time. Nonetheless, give me a few minutes and I will go write some quick comments on your BSIP.

5
Due to recent votes by large proxies, this worker is currently not being funded. Early yesterday, the 400k refund worker was voted in, which de-funded nearly all meaningful worker proposals, including this one.

Without funding, this research cannot continue as I will be required to find another funding source for the student. There is no way around this; so far, the worker has not even collected enough funds for half a year of student funding, far below the minimum required to ensure the success of the project.

If you support this research, it is imperative that you set your voting proxy to a voter who is supporting this worker and is not supporting the 400k refund worker.

Examples of proxies who are supporting this worker and not the 400k refund:

- xeroc
- evangelist-of-bts
- clockwork
- fractalnode
- bitshareseurope
- bitspark-delegate
- dls.cipher

Even now, I am considering withdrawing the research paper that I submitted to Decentralized 2019 because I may not have the funds to travel to Greece to present the research.

Dont withdraw nothing and hold on for a week or two, we are all working hard to bridge gap over this huge hole of miss-understanding and get parties getting along well. I'm pretty confident that both price of BTS and the worker will work out well for you, if not - ill personally setup and submit "additional milestone" worker for the missing funds.

As always, thanks for your continued support, and I won't withdraw the paper yet. If the BTS price recovers, then all is well. Otherwise I'm open to figuring out some kind of alternative funding source/mechanism.

6
Stakeholder Proposals / Re: Proxy: bitcrab - make the ecosystem grow
« on: August 13, 2019, 03:50:26 pm »
From my side this is not a permanent action, perhaps it will last at least several weeks.

Without certainty regarding funding for the 201907-uccs-research-project worker, I will soon be forced to move the PhD student to a different project and cease our development of simulations and analysis for BitShares. It costs approximately $37k-$40k to support the student for one year (this does not include travel funding and any funds for myself), but the worker so far has only collected about $15k.

What are your requirements for reversing your votes for the research project and the refund worker? I'd be very happy to provide more information about my research project to help change your mind (and the minds of the cn-vote team).

7
Due to recent votes by large proxies, this worker is currently not being funded. Early yesterday, the 400k refund worker was voted in, which de-funded nearly all meaningful worker proposals, including this one.

Without funding, this research cannot continue as I will be required to find another funding source for the student. There is no way around this; so far, the worker has not even collected enough funds for half a year of student funding, far below the minimum required to ensure the success of the project.

If you support this research, it is imperative that you set your voting proxy to a voter who is supporting this worker and is not supporting the 400k refund worker.

Examples of proxies who are supporting this worker and not the 400k refund:

- xeroc
- evangelist-of-bts
- clockwork
- fractalnode
- bitshareseurope
- bitspark-delegate
- dls.cipher

Even now, I am considering withdrawing the research paper that I submitted to Decentralized 2019 because I may not have the funds to travel to Greece to present the research.

8
Excellent, the paper is submitted and available for community review here.
I like your paper!

Thanks! This iteration of it is pretty simplistic, but I'm working with the student to build up a more comprehensive simulation framework and then we'll be able to say much more.

9
have gone through the paper without checking exactly the derivation of the formulas, just try to find what sense does the conclusion make.

it seems:

it make sense to set MCR<1.53, to make the optimal collateral ratio "snaps" to the right.

while MCR<1.53, then either MSSR=1.01 or 1.005 make little difference.

anyway, as the conclusion based on a simple model, so we need to consider more factors to evaluate the effect while changing these parameters.

It depends on what you're trying to do, and you're right - more study is needed.

If your goal is to incentivize low collateral ratios (more borrowing), then you need to set MCR low enough (in the paper, less than 1.53) to get that to happen.

However, low collateral ratios intrinsically increase the risk of undercollateralization, which is something that should probably be avoided.

On the one hand, low MCR should lead to tighter pegging; on the other hand, low MCR should lead to more risk of undercollateralization. This suggests that there is a tradeoff between tight pegging and solvency, which needs to be studied further.

For the research, we're starting to work up more-expressive models that should help capture these tradeoffs more clearly.

10
I convinced cn-vote to vote for the worker.

Many thanks for your support!

11
In what way/form could I give feedback on the paper?

Feel free to email me at philip.brown (at) uccs.edu.

12

As I said before even worker was around - PROPER ACADEMIC RESEARCH can only bring BENEFITS! Please let us know once draft becomes final and when we can(if we can) upload it to bitshares.org under Documentation (with full source)


We're excited to keep it going!

I will let you know about hosting it on bitshares.org; each publisher has different rules about precisely how various papers can be hosted where. However, anywhere I publish should allow full version of the paper to be hosted on https://arxiv.org/, which I will do if this paper is accepted by the Decentralized conference.

13
The first research paper is submitted for publication and available for community review here.

14
Greetings Alex, is this still on track? I just submitted a draft of a research paper for the academic track. If it's accepted by the conference, I'll plan to try to travel over there to give a talk on the research as long as my research worker proposal has been funded sufficiently by then to support a long-term project.

If it happens that I don't have sufficient levels of project funding by then but my paper is accepted anyway, will your worker be able to reimburse my travel to Greece for this? If so, the final published version of the paper would still reference BitShares as the source of funds for the work, if that's any concern to you.

Yes still on track.

I'm sure that wont be a problem...So get packing :)

Excellent, the paper is submitted and available for community review here.

15
I've submitted a draft of a research paper to the Decentralized 2019 conference in Athens, Greece (the same one that is sponsored by clockwork's worker proposal). If the paper is accepted by the conference's peer review committee, it will be published as the first BitShares-sponsored academic research, and I will travel to Greece to present the research at the conference. I'm providing a link to a draft of the research paper, but before I do, let me clarify a few things:
  • The paper considers an extremely oversimplified model of the BitShares system. The reason for this is threefold: 1) I wrote the paper in only a few weeks, so time constraints limited the amount of detail I could include. 2) At the start of a project, you really don't want to draw the wrong conclusions just because you rushed an overly-complex model to press. 3) A simple model often illuminates interesting points and helps guide the way forward.
  • One issue that the paper illuminates is that the BitShares incentive mechanism is probably very sensitive to our choices of MCR and MSSR, in the sense that sometimes if you change one of these parameters a little, it could have a huge effect on the behavior of BitAsset shorters.
  • The specific numbers used in the paper (MSSR= 1.005, 1.01, 1.02 and MCR= 1.4, 1.5, 1.6) are to be taken with a grain of salt. I used them to illustrate the general shape of the problem, not to draw any specific conclusions about those particular parameter choices.
Having given those caveats, I welcome your comments on the paper draft itself, available here.

Finally, the worker proposal which will fund ongoing research in this area needs a few more votes! Please vote for worker 1.14.204 or ask your proxy to vote for it.

Pages: [1] 2 3 4 5 6 7 8 ... 59