VELAS OVERVIEW

Velas is a blockchain platform which utilizes neural networks that are been managed by artificial intuition to boost its consensus algorithm.

The major function of the neural networks are basically to get the actual nodes operators rewards and to also ascertain block formation time. There exist the neural network weights also known as the matrix calculation servers which are found on the network members nodes for rewards generated. Large computational resources is required by the nodes in order to be able to get the actual matrix.

Tentatively, in training a neural network, genetic algorithm is required for the pre-training period and for getting the least objective function, method of back propagation of error is been utilized.

VELAS ARTIFICIAL INTUITION

Artificial intuition is defined as a set of algorithm that are used to indicate relationships and patterns in a group of data. Technically, the Velas system are made up of
Transactions per second. Usually greater them 30000. >30000
Blocks per seconds. This depends on the AI algorithm calculation. Usually between 1 second to 2 minutes maximum.

Additionally, the block time can be determined by the network load known as the TPS. If the transactions per second are many in a network, there will be a small time block. And there will be a long time block if there are no transactions in a network. And a situation whereby there is an empty block, there will only be a block header with no body whatsoever.

WHAT ARE VELAS REWARDS

A reward is given to a node generator for active involvement in the system and is based on the number of points earned. Here, algorithm changes occurs constantly and it involves:
Time per each VelasCycle
Time per each block
Transaction per each block.

REWARD COIN

A reward for block is always at 100 percent. It is shared between the participants based on the percentage of the points they earned in the neural network.

A linear regression model is used for AI which is been trained by Stochastic methods. However, the AI involves two models: Multidimensional linear regression model-This is used for the calculation of a class probability distribution density. In generating a proper decision, Bayes classifier that uses a posterior maximum estimate is used.

However, due to the existence of a super-complex environment, couples with a large inflow of data, artificial intuition based on genetic algorithms is adopted because it can handle more time efficiently compared to the neural network that uses back propagation error method. Nonetheless, genetic algorithms are used in solving the optimization problem by electing the most appropriate solution from a group of available solutions.

Alternatively, genetic algorithms for training neural networks can be adopted instead of the back propagation method of errors. The main reason for this training or teaching is to simply reduce the cost function. Plus the fact that using a genetic algorithm helps to prevent the cost function in a local minima.

Finally, it should be known that back propagation error method is faster compared to the genetic algorithms because it checks all the available solutions before selecting the appropriate one from the group of solution provided. One major setback of the back propagation method in local optima is that it involves jamming. Hence, the genetic algorithms is more suited because it’s innovative and a more preferred method to learn neural networks.

VELAS DESKTOP WALLET

The Velas desktop wallet permits the users to store their assets on their assets directly on their computer wallet . The desktop wallet is highly secured which prevents any attacks like hacking, phishing, IP spoofing, and XSS. Velas supports Linux, Mac OS and Windows.

FOR MORE INFORMATION VISIT

USEFUL LINKS
Website: https://velas.com/
Blue Paper: https://velas.com/VELAS-Blue_Paper.pdf
Technical Paper: https://velas.com/VELAS-Technical_Paper.pdf
Telegram: https://t.me/velascommunity
Twitter: https://www.twitter.com/VelasBlockchain/
Facebook: https://www.facebook.com/velasblockchain/

AUTHOR’S DETAILS
Bitcointalk Username: chicky213
Bitcointalk Profile: https://bitcointalk.org/index.php?action=profile;u=2315939