NEW TRENDS IN A.I. APPLIED TO STOCK MARKET
Usually nobody can predict stock market trends, due to almost infinite number of affecting variables and given that the finance market is intermediate between a deterministic and random system, and that you cannot model the noise, – then it is the ability to take money away when the greatest opportunity presents itself, which should be focused.
Nowadays, AI uses computer algorithms to replicate the “human ability to learn and make predictions”. AI software needs computing power to find patterns and make “quasi-human” inferences from large quantities of data. The two most common types of AI tools are called “machine learning” and “deep learning networks.”
Of course these inferences must be better than those human.
Let me cite this interesting article:
It explains how Wall Street investors, when they looked at AI models, they found that by using machine learning they can number “crunch millions of data points” in real time and capture some of the correlations that traditional statistics models (and humans) could not capture.
At the moment, the trendiest AI software is a new line of products which are conceived not only for big financial institutions, funds and investors, but also for individuals.
At the same time, the new trend in computation is the use of massive distributed computing systems based on parallel operations of many single computers or computation unities.
These unities create a network, and messages connect them in such networks (messages passing by through nodes and hubs of other computers).
For a plain description of the concept see:
This way, distributed systems are groups of networked computers, which have the same goal for their work. Nevertheless, it is possible to roughly divide different systems as really “parallel” or “distributed” using the following criteria:
- In parallel computing, all processors may have access to a shared memory to exchange information between processors.
- In distributed computing, each processor has its own private memory (distributed memory). Information is exchanged by passing messages between the processors.
We can well understand, therefore, that a computer scientists must know the math of network topology in which each node is a computer and each line connecting the nodes is a communication link.
We can forecast that the better network, according to the general theory of graphs and networks (Strogatz), will be intermediate between an ordered and chaotic system of links and nodes.
The research is going on. See a list of projects here: