Author Topic: A Call to the Truthcoin Prediction Market  (Read 21611 times)

0 Members and 1 Guest are viewing this topic.

Offline Bitcoinfan

  • Sr. Member
  • ****
  • Posts: 240
    • View Profile
These are all great questions that are addressed in the whitepaper.  I will ask Paul if he would like to respond to your questions here. 

Offline bytemaster

Ok, this write up was very dense to process given I have a headache.   Statistical analysis was not my strong suite and SVD is not something I have done in a long time.  I cannot comment on the viability of the idea from a mathematical perspective.   I can say that the write up is heavy on marketing / sales and I can say that I want to understand the exact algorithm and application of SVD. 


Quote
“Firstly, the system does not use 'voting' to establish consensus, the consensus algorithm uses a SVD-modified weighed-vote for coin-owners only. Coin owners have the highest ROI when future trading is maximized, a proof-of-stake system. The most significant multivariate-outlier-voter loses the most coins in the following period, and has the lowest relative influence within the current period. When voters lose in this way, honest voters gain, so every single actor has an incentive a] to vote honestly, and b] to lie about their voting intentions, discouraging cartels. The (required) accumulation of several votes into a Ballot is also a powerful decentralizing factor.”

Each submission is akin to mining, strengthening the overall prediction market and prevents someone from cheating or frauding others.

Please elaborate on 'akin to mining'... from what I see being weighted by shareholder balance means that a big fish can manipulate the system and cause the honest fish to be punished.   I also see a challenge of getting the data in a timely manner, especially when it isn't possible for the network to keep votes secret or to broadcast them all at the same time.   You would require a 2 phase process with collateral commitment... phase 1.. submit hash of your vote + collateral, phase 2 claim your collateral by revealing your vote in a timely manner.   This multi-phase process will generate a lot of overhead on the network and ultimately people still have to produce the feed!

In a sense, you have replaced my suggestion of a 'trusted data feed' that 'no one would want to produce' with many feeds that you expect people to produce.

I think your assumptions that 'participants want the value to go up' and thus will tend to be truthful may ignore certain incentives to gain personally even as it goes down.

Here are the challenges I see:
  a) With SVD you need to keep peoples individual votes secret and this is a challenge for a DAC
  b) Someone with a large stake can swing the consensus and pull everyone to his position.
  c) On a DAC there is no way to recognize individuals and track reputation (required to punish them in the future)

Quote
In any case, why pursue the message of decentralization, if we’re compromising on centralized feeds?

I think I need to clarify something very important.   Decentralization is a means, not an end.   I want to create an entire video addressing this concept because decentralization is kind of a religion that has taken on a life of its own and become an end to itself.

Lets look at the problems we are trying to solve with decentralization:

1) Single point of failure. 
2) Single point of control.

These two problems have varying degrees of severity and depend upon the threats you face:

1) Is it failure from accident, malice, or attack? 
2) What is the nature of the control?  Absolute, Finite, Limited?

If a centralized solution can perform the job, eliminate corruption, has low probability of government shutdown, and have redundant backups in case of acts of God or government then it will likely serve the needs of the market more efficiently than a decentralized attempt at the same problem.

From an economic perspective you must analyze what kind of market depth (participation) is required before your SVD approach can be effective.  If there are only 3 participants (long/short/feed) then I believe the sample size would be too small.   However, add a bit of trust into the equation and it would be fine.

Addressing your Concerns with my approach...

1) If there is demand for a feed people will be willing to pay enough money for the feed to exist so they can bet on it.
2) Many companies can make good money producing feeds
3) A simple 2 out of 3 judgement would be enough to prevent any one feed source from intentionally tanking the market.

I think I would like you to walk through the role of each actor in the system, how they get their role, how / when do they produce information, etc.

I am very happy to work with you on this idea... but you need to get it down to a simplified explanation that can allow us to crowdfund this idea.

For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Offline Bitcoinfan

  • Sr. Member
  • ****
  • Posts: 240
    • View Profile
A Call to the Truthcoin Prediction Market

To Bytemaster:

Let me start off by saying, I am a huge fan of what you are doing.  It’s clear to all of us on the forums, with each update you make, and each new development posting, each effort to answer a new member’s question, you are proving to us your commitment to the success of Decentralized Autonomous Corporations.  You are staying true to your creed of, “Creating something great that will be here long after we are gone.”

Executive summary

The Truthcoin project is seeking to work with Invictus Innovation as a collaborative project that will be synergic (non-competitive) with other companies in the DAC pipeline.  Truthcoins is a Decentralized Prediction Market Protocol that uses Singular Value Decomposition, a consensus correlation amongst Truthcoin holders at the maturity of a Decision Contract.  The correlated votes of Truthcoin holders determine True and False payouts for any matured Decision Contract.  Truthcoin Prediction Markets (TCPM) can act as insurance contracts, inventing new means for individuals to profit from their creative work, without restrictive legalities of patents and copyright.  A Decentralized Prediction Market can also function as an Executive Operational Management System for Distributed Autonomous Corporations.  Participants in DACs can create markets to predict the return on accepting an investment venture, or manage the hiring/firing of staff in a DAC in the same manner Human Resources or a Board of Directors would do in a Chartered Entity.  A last practical use of Prediction Markets is the ability to support crowdsourcing, micro lending by establishing a market for creditworthiness.  Publicly Funded projects like bridges and roads can be performed without the need for funds from municipalities.  Incorporation of Truthcoins into the Keyhotee Application (similar to Bitshares X) would support the DAC ecosystem. 

These new opportunities in predicting and hedging risk in any sector, would match the global success of the derivatives market, which is upwards of a $791 Trillion Market Capitalization.  Truthcoin holders would have a stake in this immense market by receiving the trading fees on the marketplace, providing the greatest wealth return for their shares. 

Introduction

As a long-time shareholder of Bitshares and an early member of the forums, I want to encourage Invictus to take a closer look into Paul’s work in the Truthcoin Whitepaper.  I personally know Paul, Asymmetric Information of Bitcointalk.org, and hold regular contact with him about his project.  I can say he is very passionate about the future that both crypto-currencies and prediction markets will bring.  Paul deeply understands statistics and forecasting.  The work with his colleagues at Yale (not a no sweat school either) and their input have allowed for his paper to be mathematically rigorous. I myself work within one of the country’s largest financial institutions in areas of predictive financial modeling and risk management.  Paul’s work is nothing short of remarkable.  This is an idea ahead of its time.

He has discussed with me the possibility of Decentralized Prediction Markets for over a year now, and we both knew this would allow for the formation of a Decentralized Autonomous Corporation—we personally had just called it “Decentralized Companies”.
We both see prediction markets as the next natural progression after Bitcoin, since it allows to people to bet anonymously (discouraging herding like on publicity traded stocks), and now because we have Bitcoin we can make financial instruments out of IDEAs.  Any idea we can conceive can become a financial asset; harkening back to the Internet’s promise of “Permission-less Innovation.”

Today, the combined global wealth of equities and stocks is upwards of $100Trillion.  $100 Trillion for a market that is highly regulated and saddled with enormous barriers to entry.  You and I simply can’t go to Wall Street and tell underwriters to IPO for us our company.  These gatekeepers will laugh, and then scold us for wasting their time, explaining that only companies in their league can afford the prestige of an IPO.  Even with gatekeepers at the door, a collective net wealth of $100 Trillion has crossed the threshold.  For a moment, think of how large a market for ideas could become.  Without the barriers to entry, the red-tape, the underwriting middle-men and empty-suit analysts, the wealth generated would be astronomical. 

You will hear me make this statement now, and continuously:  Invictus should look to implementing Truthcoins, because each AGS PTS owner will profit from it.  The Truthcoin holders will be able to earn on each and every single trade on this market through its consensus mechanism.  PTS and AGS holders, will earn profit from ballot submissions on each contract.  Some people on this forum will be able to make careers out of this.  Who could object to that?

Why Truthcoin protocol?
a) Singular Value Decomposition
Paul has prudently chosen the most favored statistical technique for evaluating the basis of consensus called Singular Value Decomposition (SVD).  SVD was discovered around 100 years ago and it can be said to be the pinnacle of linear algebra.  SVD is a useful statistical technique that has found application in: engineering bridges, measuring IQ tests, data mining large populations of research, aircraft design, gene data analysis, seismology, animation and digital imaging processing.  It provides the highest possible correlation, on lowest possible factors and is ubiquitously well regarded.  In a more recent cases, SVD was used to verify that Dorian Satoshi Natamoto’s writing style was not that of Bitcoin’s Satoshi Nakamoto. 

Here is a quote from Paul as to why SVD is not the equivalent of democratic voting:

“Firstly, the system does not use 'voting' to establish consensus, the consensus algorithm uses a SVD-modified weighed-vote for coin-owners only. Coin owners have the highest ROI when future trading is maximized, a proof-of-stake system. The most significant multivariate-outlier-voter loses the most coins in the following period, and has the lowest relative influence within the current period. When voters lose in this way, honest voters gain, so every single actor has an incentive a] to vote honestly, and b] to lie about their voting intentions, discouraging cartels. The (required) accumulation of several votes into a Ballot is also a powerful decentralizing factor.”

Each submission is akin to mining, strengthening the overall prediction market and prevents someone from cheating or frauding others.  This forced feedback system cannot be manipulated by a cartel of voters. 

b) Multi-Dimensional Markets
Byemaster has proposed for event based prediction markets, a system of competing data feeds, whose credit trust is based on market verification.  Those who fall below a market threshold price will not be included as a prediction market feed.  This concept has merits of its own, and I’m not against this type of validation—however the real question we should be asking is why is this centralized data feed market competitive at all?

Exporting data feed is a horrible business to be in. There are no reasons why you would want to be doing this. What would prevent someone from providing a good feed today, to accrue good standing with the market, and then sabotage it later on?  What’s the incentive of providing an honest feed?  The work is tedious and dull. There is no scope for consequences of your wrongdoings.  If the incentives are small, how will the market be competitive?  In any case, why pursue the message of decentralization, if we’re compromising on centralized feeds?
Additionally, Truthcoin strives to improve upon the primary idea of Prediction Markets with an idea called a Multi-dimensional market.  That is to say, you can create winner take all decision scenarios.  For example, as in Paul’s example, he has a large 4 by 4 matrix of the chances of the 2 2015 Jobs Act” Passed and Signed, and the chances of Unemployment <6% for April 2015.  (Look to Examples in  #3 in https://github.com/psztorc/Truthcoin/blob/master/docs/2_PM_Types.pdf

In a way, a multi-dimensional system, allows us to mesh together hundreds upon hundreds of variables, each time, revealing the risk with each additional one.  The system provides granularity.  This will be more accurate and offer greater precision to the narrative of what is actually occurring.  Today, as in the case Bitshares Prediction Markets, we can only doing statements that are single faceted, without room for enhancing with additional variables. 

Because Truthcoins, permits the ability to attach infinite possible scenarios, the features of multi-combinatorial decisions would be too cumbersome to implement with validation from centralized data feeds.  It appears counter-intuitive for a competitive market feeds to verify the exponentially growing and limitless combinations of just one decision.  With Truthcoins, we would like to defer to the Truthcoin holders, since they should act rationally as to maximize the future value of their coins, if they choose to sell it, or profit from the income streams.   

More examples on Multi-Dimensional Decision Contracts look at  https://github.com/psztorc/Truthcoin/blob/master/docs/2_PM_Types.pdf

The Benefits of Prediction Markets
a) Example Problem: Media Proclamations

Talking heads rule our perceptions.  Radio hosts, media outlets, blogs, and newspapers, what have it; inaccurately define public reality.
Their talk is convincing, because they are a tuned to the left hemispheres of our brains—the processing language centers. They have us believe that something is true, especially when they place importance on technicalities.  What the media doesn’t see has fitting their models, is convincingly cut off.  It simply does not fit.  What becomes, is an improper selection of information.  Thus we get into a trap. Paranoia and dramatized sound bites become prevalent; myths and pop-science flourishes.  The sensational and scandal list grows tall.  The dumbing down of scientific research for the media consumption—such as, research on serotonin has been shown to imbalance has been shown to cause depression—would allow media and their marketers to say new studies show eating ice cream and grilled cheese sandwiches helps fight depression. 

In our community this misinformation was evident when the famed Silk Road Dread Pirate Roberts was arrested by the police with some news outlets reporting things as egregious as “The creator of Bitcoin, Satoshi Nakamoto funded Dread Pirate Roberts”.  In our current left-hemisphere world, headlines proclaim all.  When these “facts” are debunked later—the damage has been done.  Prediction markets are needed to rein in this sensationalism. 

b) Solution: Empirical Consensus

Truthcoins Prediction Markets solves this at the core.   What Prediction Markets offer is a determination at what % these stated facts are True or False.  Prediction Markets can put a stop to these endless arguments.  If you disagree about something, make a Prediction Market for it.  Put your skin in the game.  In a post-Truthcoin era, those who make severe proclamations (such as the likes of Paul Krugman, Mark Williams, and Warren Buffet who see no value in Bitcoin) will be called to put up or shut up.  In this world, to be taken seriously, they should put their money and be held accountable to when their right or wrong.  We can say, “If they don’t, then why should their advice be heard?  If you, yourself aren’t going to put it on the line for what you believe, why should I?”

The same idea can be brought here on these forums.  On any forums, the loudest minority opinion can influence actions, thereby clouding the perceptions and understandings of new DACs.  Why deal with them and their arguments, when one can simply look on the prediction market to see empirically what consensus says.  By what percentage is it better to accept this idea?   Bitsharestalk.org, in more recent weeks, has had disagreements about what is the next single-best DAC.  Invictus can solve this problem by posing the question—Do you think DAC XYZ, DACABC, or DACPSP will produce the highest ROI by 2014?  Its easy to ask those vocal members on the forums (our future leaders), to publically reveal their bets made on the Truthcoin blockchain.  Their opinion, and the stake the hold in that opinion are symmetric.  From here we can see how this will allow Invictus and other DAC entrepreneurs to make a decision: the highest performing DAC on the prediction market should be adopted because the results show it will succeed.  The prices will contain the information of which DAC the community/ market believes will have the highest future value.

Low-cost Hedging for Research and Development

This same concept can be adopted to understand the future benefits of any investment project.  This has a profound impact in the R&D space.  A company or organization can create a Decision Contract with the statement:  The Invictus Mobile Phone will be the highest marketed phone in the United States by the end of year?  Or, Will the Invictus Mobile Phone exceed sales of $2MM for the year 2020. 

Non-Profit Organizations can create statements such as: Company’s X will develop a new product that will eliminate the cold virus by the end of 2035?

A company such as Tesla can hedge their bets on certain projects by creating statements like:  The new driverless Telsa will exceed X units globally in the year 2021?  Then Elon Musk can buy up shares that Tesla’s new driverless car will not exceed this many units for purposes of reducing downside risk.  Hedging will allow companies to take advantage of the high growth of their investment projects, and dramatically reduce the fallback if the project were to be a failure or below expectations.  Its an effective, low-cost insurance within any playing field. 

Patents and Copyright Material

Bytemaster has already touched on this several times.  Prediction Markets will bring in a new system of principals that will eliminate the need for patents and copyright laws.  Within the movie, music, and book publishing industries, Prediction Markets will allow the original authors to benefit from consumption of their work.  It will also reduce the costs of distribution by eliminating those “gatekeepers”.  For example, a new 15-year high school author, who has finished writing her book, has been having a hard time finding a publisher in the traditional markets.  They all don't like her book for whatever reason.  With a Truthcoin Prediction Market, the young author can upload his book onto whatever distribution channel she wants (decentralized Mediashares or centralized Amazon/ Apple bookstore) and write a Decision Contract that will say: “This book will exceed 500K purchased downloads on site XYZ by the end of June 2014.”  The young author can then purchase shares of Yes on the Truthcoin Prediction Market.  Afterwards, it becomes a win-win situation for the author and reader.  Any casual reader can download her book at low cost (.001 BTC micropayment as a spam filter) and the author can directly profit from those enjoying her hard work.  Bytemaster introduced this concept a few months ago as AP Shares, where there will be tradable shares for an article’s value.  This can also take place with Truthcoin, and is a much more simplified implementation, allowing the young author to underwrite and IPO tradable shares herself for people to buy, as opposed to a Yes or No Statement share.  These are just a few of the limitless possibility of the Truthcoins design in the Copyright and Patent field. 

Executive Management System for DACs

The Father of Prediction Markets, Robin Hanson spoke about the power of prediction markets to estimate the performance of CEO’s.  A PM can track the CEO’s performance with the statement: Will CEO XYZ still be in the job at the end of Q2 2014.  Yes and No will display his reputation.  Truthcoins are the beginning of a DAC Operations Platform.  For a DAC, the platform will have markets to self-regulate itself.  There will be an endless host of statements for the performance of the company

(See example 4 of Paul’s 2_PM_Types Github https://github.com/psztorc/Truthcoin/blob/master/docs/2_PM_Types.pdf). 

Statements will included: Will the CEO, VP, Marketing Director, Product Designer and Developer, COO, Supply Chain manager, or middle managers be fired at the end of 2015?  Whatever percentage results, will clue into the performance of that employee.  A high marginal percentage means that employee is doing their job properly.  And a low would be high chances unemployment.

This improves the current shareholder voting system we have in the traditional financial markets.  Smaller shareholders have little input overall, and if they did, it would be in only the trivial issues such as changing the guards (CEO’s) whenever the Earnings per Share are low.  Prediction Markets will allow those invested in DACs to macro- and micro manage their decentralized company.  It will numerate all types of risks across the board: Operational, Business, Political and Legal, Valuation, Project Risk, etc.  Transparency becomes ubiquitous at all levels, thus pushing the envelope for how DACs should be structured.  The markets, through prediction markets, will manage the talents and resources within Distributed Corporations, and make them effectively autonomous true to their value.

From here anybody can see what consensus believes is the best choice.  And unmanned companies can maneuverer accordingly.  Remember this is not voting consensus, but market empiricism.  Those who don’t believe the idea will work or the company will follow through.  This is the footprint for what DACs are to become.   


Prediction Market Based Reputational System

This should already bring thoughts of a Reputational Based system incorporating PM’s sometime in the future.  For example Credit scores and Probabilities of Default can all be extracted from a Prediction Market.  Today we rely on Credit Rating agencies such as S&P and Moody’s for their expert Credit Ratings for countries and corporations.  But humans are fallible.  We all saw in the 2008 crash that Moody’s had turned a blind eye to dangerous leveraging that was happening at the time.  The United States today still has an AA+.  Do we think the chances of the United States defaulting on its promises to be that low?

For the ordinary consumer, adopting Prediction Markets hold great promise for P2P lending, by sorting out that friction and high human overhead required for securing a loans from Financial Institutions.  It can be done with the statement: Will this person or small business get a loan at this time frame?  The percentage of Yes’s derived will say what the consensus thinks the users Credit Worthiness is.  This will gain traction within new businesses and DACs who want to create global lending clubs.  Community Based Development, publicly funded projects -- profit and nonprofit, i.e. Repairing or building a town bridge, would benefit immensely with this free market lending environment.  PM will be the roadmap to decentralized Governance, such as a Futarchy system

(What is futarchy -- http://hanson.gmu.edu/futarchy.html).

(More interesting PM applications https://github.com/psztorc/Truthcoin/blob/master/docs/3_PM_Applications.pdf)


Keyhotee and Prediction Markets

In our eyes, Keyhotee and TCPM are complementary.  TCPM benefits from Keyhotee’s user interface and multi-application wallet.  TCPM would be able to have Bid/Ask capabilities just like Bitshares, and can enable end of Decision Contract ballots on the already existing username system of Keyhotee.  Users can login into their Keyhotee account and periodically vote if a matured statement was correct or not. 

Over time, five years after the start of TCPM, we will be able to look through the blockchain and see which usernames were the strongest voters—the “experts” in their respective Truthcoin fields. The Truthcoin blockchain would be also a free, open-source, consensus verified dataset for everyone, from academics, to analysts, to researchers, students, to decision-markers to explore.

For Keyhotee, TCPM would open doors for a DAC platform—a means for operationally managing an unmanned DAC.  Keyhotee would establish itself as the iTunes of digital financial instruments, the application library for crypto-equities, i.e. small business IPO’ed coins, Movie IPO’ed coins, Bookcoins, Articlecoins, Pop Artist coin, etc. 

I heard enough!!!  What’s in it for us?

Since TCPM become equity derivatives of every idea in the world, the return on far surpasses a Billion-Dollar Industry.  We already know derivatives within current legacy banks dwarfs the size of the world stock market by an estimated $791Trillion to the 36.6 Trillion of global stocks.  Every PTS and AGS holder who votes in the system will get a piece of this pie. 

When implemented, prediction markets have the chance to impact every sector, industry, organization, political and education system in the world.  Built in a decentralized protocol, its adoption will be dramatic.  I strongly encourage Invictus to sponsor this project, as the payoff truly maximizes shareholder wealth. Paul has designed the TCPM in such a way to play a deep domain role within unmanned DACs.  The TCPM protocol is built on incentivizing security, proof-of-consensus, and builds on the very same notion Bitshares, by decentralizing the approaches to the problems that confront us in a top-down heavy paradigm.  Like the insurance contract idea, many early PTS, AGS holders will be able to make a career out of proof-of-consensus voting.  Truthcoin holders will have a steady stream of income from performing the required bi-weekly vote, their earnings coming from the pool of fees collected from trades in the decision market.  The more Truthcoins you have, the greater you’re earning power.

Remember, this is a market of ideas.  The only constrain is our imagination.  A resource that is infinite.  That is to say, as John Muer put it. “The power of imagination makes us infinite.”
For us, the powers of prediction markets are unbounded.  For a full read on Truthcoin Prediction Market look at: 

https://github.com/psztorc/Truthcoin/tree/master/docs

https://bitcointalk.org/index.php?topic=475054.0

-B
« Last Edit: April 05, 2014, 09:49:45 pm by Bitcoinfan »