EFFECT.AI THE DECENTRALIZED NETWORK FOR ARTIFICIAL INTELLIGENCE

Artificial intelligence is intelligence demonstrated by machines in contrast to the natural intelligence displayed by humans. In computer science artificial intelligence research is defined as the study of intelligent agents. An intelligent agent is any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goal. In lay man terms it means when a machine makes smart decisions so as to make it look like that of man. Lets say I have a music system that could determine my mood and play a song to match my current mood you can call that artificial intelligence because the system would have learnt over time what music I play when am in a particular mood and by playing songs that match my mood has solved problems,since the aim of artificial intelligence is problem solving.
The Artificial Intelligence (AI) market is growing at a remarkable rate but has become more inaccessible than ever. The requirement for large annotated datasets and a complex technical infrastructure has driven AI development behind the closed doors of corporations. Allow me introduce a decentralized network called Effect.AI, that provides services in the AI market.

One of the biggest problems currently facing the development of AI is the fact that most algorithms are developed by large corporations. As a result of this, there is reduced operability between various algorithms, making the AI environment as a whole non-cooperating. People who need results or data sets pertaining to these algorithms struggle to find it or end up having to pay a premium for it. People who want to self-train AI also contend with these problems and, overall, it is a very expensive endeavour.
To tackle this problems effect.ai would make use of a decentralized mechanical turk powered the NEO Blockchain called the effect network.

what is a mechanical turk?
a mechanical turk is a crowdsourcing Internet marketplace enabling individuals and businesses (known as Requesters) to coordinate the use of human intelligence to perform tasks that computers are currently unable to do. Employers are able to post jobs known as Human Intelligence Tasks, such as what song should be played when in a happy mood. Workers can then browse among existing jobs and complete them in exchange for a monetary payment set by the employer, In this case the payment is NEP-5 token. The effect network is three phases.

Phase 1 : EFFECT Force
The first phase of The Effect Network will be an interactive marketplace for tasks that require human intelligence. A Decentralized Mechanical Turk called Effect Force. It allows anyone in the world to perform a wide range of tasks and receive fair payment it would be like fiverr on blockchain, sounds interesting right? It will give AI developers and businesses access to a large workforce of human intelligence to train AI algorithms. When a worker completes a task, they are paid with with a network NEP-5 token. so for a company that wants to perform my dream music system that plays music based on my mood they would have to collect data from several people.Now lets leave my music system other tasks that would be solved by the decentralized on-demand, scalable and distributed workforce include image classification,audio transcription and data labeling.

Phase 2 : EFFECT Smart Market
The Effect AI Smart Market is a decentralized platform where people can offer and buy AI services. This phase is a natural progression of the network. It will be an open marketplace for offering AI Algorithms as a service. This marketplace or exchange is where AI developers with a functional Algorithms can sell, rent or give out its service for a simple and easy payment with the Effect token. Each Algorithm will have its own wallet to allow for easy acceptance of transactions. Also within this marketplace, Algorithms have the ability to communicate and collaborate with other algorithms and and purchase services from each other.

Phase 3 : EFFECT Power
The last phase provides a decentralized, distributed computational platform that will run popular deep learning frameworks. The Effect decentralized compute engine is based on popular deep-learning networks like Caffe, MXNet and Tensor flow. We want to distribute the computational power of all AI so there is not this one point of weakness. We can do this through partnerships with projects like Golem with their Super computer or create the framework ourselves.

How can effect force help employment in developing countries?
Unemployment is one of the highest socio-economic challenges that are negatively affecting the world at large. It is more detrimental in developing countries, the effect force would grant source of income to millions of unemployed people especially youths and can also serve as a means of alternative income to those who already have jobs

How can fair and instant payment impact adaptability?
A large number of workers on Mechanical Turk are mainly young people from the United States and India, purely by design since tasks can only be posted by those with US bank accounts, and money could only be transferred to those workers with US or Indian bank accounts. This probably explains the lack of contributors from other developing nations. However, since workers would be payed instant tokens at a fair price it would encourage more people to join the work force because they wont need to have a us bank account to withdraw their earnings.

problems that micro-tasking can tackle

1. It would help to provide various solutions to a problem in the quickest possible time
2. It would also help to break down large task into micro tasks
   3.people who need AI data no longer need to go through third parties, thereby cutting the cost and reducing the long wait periods for verification and payments

Practical example of problems that can be solved using the Effect Network

1. Say I need to classify road signs in images. This is a crucial task when it comes to self-driving cars for instance. AI algorithms first need high quality data that is properly labeled to learn how to recognize each road sign. This is Phase 1 in the Effect Network: a large workforce of humans can classify the road signs in an image data set.

2. After a sufficient amount of data has been fed to the algorithm, I have an AI that can successfully classify road signs in images and I want to put it to use. In the second phase of the project (Effect Smart Market), this algorithm can be listed on a marketplace where users and other applications can easily find it and compensate the developer for its use. Every someone uses the algorithm I have listed, I will earn some EFX.

3. To provide users access to the algorithm it has to run on a computer somewhere. Right now, this is still being done using centralized means of processing. The more the algorithm is used, the more power it consumes. In the third phase, Effect Power, this computational cost will be distributed over a global network.

For more information, check out:
Effect.AI Website
Effect.AI Whitepaper
Effect.AI Lightpaper
Effect.AI Youtube
Effect.AI Medium
Effect.AI Facebook
Effect.AI Twitter
Effect.AI Telegram
Effect.AI Review/Explanatory Video

This is a contest sponsored by @originalworks. You can check out the original post here

effect2018