Machine Learning and how CryptoAngel is going to improve it
What is machine learning, and how CryptoAngel improves it?
Machine learning is a subfield of artificial intelligence that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. It focuses on the development of computer programs that can learn for themselves, given the collection of examples (data) and the goal that needs to be met.
It is divided into supervised learning and unsupervised learning. In supervised learning label data is given to the machine, which then uses this data to build the model of it and use it on newly seen data to make predictions or label them. The example of this is image recognition, where machine can learn how to recognize images and output what it sees (for example: is it a dog or a cat). In order to build a model, we need to provide it with a lot of training images of dogs and cats (explicitly saying this is a dog, that is a cat). The other technique is unsupervised learning, and it operates on raw data (not labeled), and is able to find hidden patterns or anomalies inside the dataset, or it can be used to split into clusters of similar point.
Machine learning is a data analytics technique that teaches computers to do what comes naturally to humans and animals: learn from experience. Machine learning algorithms use computational methods to “learn” information directly from data without relying on a predetermined equation as a model. The algorithms adaptively improve their performance as the number of samples available for learning increases.
Machine learning algorithms find natural patterns in data that generate insight and help you make better decisions and predictions.
Deep Learning as one of the most efficient machine learning techniques
Deep learning is a particular kind of machine learning that achieves great power and flexibility by learning to represent the world as nested hierarchy of concepts, with each concept defined in relation to simpler concepts, and more abstract representations computed in terms of less abstract ones. It is based on neural networks, which are the artificial representation of human brain. Neural networks are composed of artificial neurons layered in many levels and interconnected resembling the structure of our brain. These neurons and connections between them are configured in the training phase based on the dataset given to it.
Deep learning refers to artificial neural networks that are composed of many layers (called hidden layers). The ‘deep’ refers to multiple layers. In contrast, many other machine learning algorithms are shallow because they do not have a Deep architecture through multiple layers. The Deep architecture allows subsequent computations to build upon previous ones. We currently have deep learning networks with 10+ and even 100+ layers.
The presence of multiple layers allows the network to learn more abstract features. Thus, the higher layers of the network can learn more abstract features building on the inputs from the lower layers. A Deep Learning network can be seen as a Feature extraction layer with a Classification layer on top. The power of deep learning is not in its classification skills, but rather in its feature extraction skills.
Deep learning is especially effective in image recognition, which is due to its ability to extract and abstract features. For example, recognizing a face in a photo has many layers of recognition: recognizing eyes, hair, ears etc., and this is where deep neural networks excel.
Problems with current centralized architectures
Training a deep neural network is extremely expensive, computation vise, and it requires huge amounts of data. So, in order to build an effective deep neural network, it often takes a week to train it using hundreds of machines equipped with expensive GPUs.
So, unless you are Google or Facebook, the infrastructure needed to build a great performing deep learning model is not available for you. One of the solutions would be to resolve to one of the cloud computing platforms, which are usually very costly (around 0.74$/hr per GPU).
The other problem that was already mentioned is where to get such a huge dataset used for training this model. These datasets are usually not publicly available, and are in possession of companies and institutions who gather them on various ways.
There is a need to democratize and decentralized AI, meaning give access to these resources to everyone who has the skills and knowledge about these topics, and only in this way AI can advance to the next level.
Decentralizing AI
CryptoAngel aims to provide ecosystem which will accelerate development of generalized and democratized AI in decentralized manner. Business ecosystem is encompassing several crucial participants that drive organic growth of our AI application, where value is exchanged by means of Angel cryptocurrency. CryptoAngel will provide a framework as a tool for external users to build AI models, provide data training sets and to contribute to the development of what we call Common AI in return for Angel cryptocurrency. Common AI is the brain of our internally built system crowdsourced by external domain (subject matter) experts. It is decentralized application consisted of neural networks i.e. AI models, and logically divided into subsets of categories where each category has value for end user. One model can be found in several categories.
CryptoAngel ecosystem will interconnect following participants:
· Data providers — owners of datasets who will be incentivized by our cryptocurrency to provide their datasets to be part of our big data training pool. We will set specific requirements for data sets through our portal, those requirements will come from classes of AI models designed for specific domains of problem.
· External developers — AI developers who will submit their AI models to be part of Common AI in return to Angel cryptocurrency. They will also have an option of training their models on the distributed network and use the miners idle computing power.
· Miners — create decentralized blockchain consensus, mining with proof of intelligence algorithm to train AI models and verify computations in CryptoAngel blockchain network. Miners will invest their computation power to receive remuneration in Angel currency.
For more information visit: https://www.crypto-angel.com/
