Are you - as a crypto trader - realizing your full potential without using artificial intelligence?

As focused as you might be on crypto-currency as the most impressive computing innovation since 2008, you shouldn't discount the potential of artificial intelligence advances.

However, if you've ignored AI, consider me in the same boat as yourself!


Why would I ignore headlines about AI?  Because even though it makes for great science fiction novels and movies, I'm a developer; my background is coding in web technologies, and I know that programs essentially "do what you tell them to do."  To me, AI was a just a fancy way to say "human-created algorithm."  But Silicon Valley didn't see it that way; instead, the field of artificial intelligence has exploded in the last five years.

Is True AI Even Possible?

The human brain contains approximately one hundred billion neurons.  That’s a large number, but keep in mind that it’s the connections within a human brain that are even more incredible.  Scientists estimate that each neuron has no less than 7,000 synaptic connections to other neurons.  Each adult human – you and me – has around 500 trillion synapses.

Another incredible feature of the human brain? It’s the original internet inside one organism; Each neuron is able to contact any other neuron with no more than six connections – or “six degrees of separation.” 1

Do computers exist that have this capability?

Distributed computing can reach these levels; however, singular machines with this capability are very rare.  Matt Mahoney, a Ph.D. of Computer Science, stated that the hardware itself exists to process 10^16 operations per second, but those machines cost millions of dollars and consume several megawatts of power.

It’s not just the hardware; it’s estimated that the human genome has the equivalent of 300 million lines of code written into our DNA.  genome In addition, it requires humans years of training to do the most complex tasks.  Paying developers to write that much code costs a large amount of money. This is why AI has ordinarily been pursued only by companies with very large checkbooks, like Google. 2

One approach to AI is to focus on large data sets and then compare new inputs to the existing set to determine a fit.  Various schools of thought label this approach with different names depending on context: Memory-Based Reasoning or Case-Based Reasoning. 3

One of the most famous AI pioneers, John McCarthy, felt that trying to replicate human capabilities was too limiting and proposed that artificial intelligence could solve problems successfully using their own approaches, regardless of whether a human would solve it the same way. 4

Current Real-Life Applications

AI algorithms contained within smartphone applications have become so pervasive that you don’t consciously think about using them anymore – they are just “very convenient” and we struggle to remember what life was like before them:
  • Language Translation
  • Maps & Directions
  • Internet Search Results
  • Personal Assistants
Most of these are founded on concepts originating from neural networks, or artificial systems that attempt to efficiently simulate pattern recognition found in nature.

AI Applied to Stock Trading and Crypto-currency

A lot of moneyed interests have been keenly trying to keep ahead on the latest abilities of neural networks to identify trade strategies for stocks and other financial assets.  Simply speaking, trading is ideally suited to AI applications, given its trove of historical data and real-time information.

Marcie Terman, head of the Communications Division of First Global Credit, had this to say about the potential of AI and crypto markets: 5

“There is room while cryptocurrency markets are still in their infancy for AI developers to create systems that learn to identify profit opportunities in these young, highly volatile markets.”
Some open-source development projects provide evidence that she's correct.  For those crypto traders looking to test their own algorithms or AI-created trade strategies, some projects have been created by enterprising coders:

Other Crypto-currency & AI Examples

Numerai is a hedge fund based in San Francisco.  It was founded in October of 2015. 6  AI algorithms perform its trades, and those algorithms are submitted by data scientists on a monthly basis, in return for the possibility of obtaining some Numerai tokens (a token created on the Ethereum network).Numerai

The hedge fund trades stock market securities, not crypto-currency, however; it uses the value of the Numerai token only as a reward for the contributing data scientists. 7

Catalyst is a crypto-currency trade platform founded by former MIT students. 8 They released their whitepaper on June 6, 2017, describing Catalyst in more detail: 9

“The main goal of Catalyst is to serve as a one-stop shop for developers (or quantitative traders) who are interested in developing trading strategies that operate in the expanding domain of crypto-markets. Developers can utilize the myriad of data sources that will be made available through our plat-form, and will be served through Enigma’s peer-to-peer data marketplace protocol, to build their models, back-test them according to historical data, as well as put their strategies to the test in a simulated or real trading environment.

We are also creating a data marketplace to drive better crypto-investment decisions. Algorithmic trading, and in particular, strategies based on AI models, are only as good as the input data fed to them. The ecosystem surrounding crypto-markets is still in its early days and relevant data sources are scarce and fragmented.”

The team is conducting an ICO to fund development of their project, scheduled to run from September 11 to September 21. 10

Rialto.AI and the XRP Ledger

Rialto.AI’s ICO from July 1 to July 14 of this year raised ten million dollars. 11  The group has a business plan to execute proprietary algorithms for arbitrage and market making.  Their approach is to provide liquidity and matching orders using AI algorithms. 12

To initiate the ICO, the team released a whitepaper 13 that described their architecture; the whitepaper caught people’s attention when they noticed that a decision had been made by Rialto to utilize gateways within the XRP Ledger:

“The greatest potential to be exploited is in creating the market between gateways within the Ripple payment protocol.”

Ripple Solutions

In addition to third parties building on the XRP Ledger, Ripple has also ventured into artificial intelligence.  On April 27th, Asheesh Birla, Vice president of product at Ripple, was quoted as saying: 14
“We have a very robust research team that is looking at ways to increase our efficiency—machine learning for fraud detection, compliance.  We want to be on the forefront of machine learning and AI in the same way we are on the forefront of blockchain.”

In Summary: XRP Ledger Use Cases & AI

In addition to market making use cases, other core use cases of XRP might also benefit from third party AI solutions and integration:
  • ForEx Trading
  • Arbitrage Trading
  • Big Data Analysis
The XRP Ledger is “open” and anybody can build these applications to study and use the data sets from the XRP Ledger history; one of the tools of modern AI is the training aspect whereby data sets are targeted for analysis and learning.

As AI becomes a standard part of trading software, expect to see more institutional fund managers asking for the full history of the XRP Ledger as they start to invest more of their assets into XRP’s future!  xrp_symbol4