EFFECT.AI || PHASE 1 - EFFECT FORCE [AI & HI COLLABORATION]

Source.

The Collaboration of Man and Machine which was once seen as mythical by many has become true as true can be. Human Intelligence (HI) which once limited the reach and growth of man's development is now being consolidated by the relatively new Artificial Intelligence (AI) and it's funny or paradoxical because a limited, error prone Human Intelligence is responsible for the creation of this seemingly unlimited, zero-error Artificial Intelligence.😏
The proximity of components required for its development still lay scattered around, with to solid and secure connection between them.
Effect.ai takes the first step in organizing and connecting all the components required for speedy development of the AI industry. Some of these components include:
• Datasets.
• AI Developers.
• A Marketplace for AI products.
• Computational Power.

Considering each of these components carefully, it becomes evident that each one of them and the interactions between them as well have contributed to lowering the effectiveness of the AI industry.
As regards Datasets, they play a crucial role in the development of the AI industry. Training a machine to make decisions for itself requires datasets, which are split into training data and testing data. Although the AI algorithms are available for specific decision domains, if the datasets used to train machines aren't big or clean enough, it results to poor performances and limited capacity.
Machines are not going to train themselves to think as well, Its a good thing they can't because If they could we would have been in trouble.🤣 This arises a need for AI developers who can optimally train these machines for the required conditions with the available AI algorithms such as Naive Bayes, Decision Trees, Support Vector Machines (SVM) e.t.c.
Associated with this component are problems of criticisms, nonavailability of marketplaces for their products, a platform for collaborative development and start-up support, in finance or computational power. A consequence of this, is an uneven Development of the AI industry around the world.
These problems may have been spelt out here in just few lines, but they are more enormous than they seem.
A well orchestrated solution for these problems, would be one which handles the problems of this AI industry separately and in-depth, a divide and conquer kind of solution because the factors that affect each of the components are notably different. Also, being the first of it's kind, to avoid autonomous control which would lead to unnecessary exorbitant fees, downtimes and to ensure security and scalability, the solution should be decentralized.

Effect.ai creates a decentralized AI network, the first ever, which aims to solve these problems of entry and remaining in the AI space encountered by AI enthusiasts and experts. Combining the power of the Blockchain (specifically NEO) to these innovative solution increases the feasibility in terms of scalability and security.
The development of this blockchain solution which will serve as a vehicle to further propel the the growth of the already rapidly growing AI industry will occur in three phases. They Include;
• Phase 1 - Effect Force(Mechanical Turk).
• Phase 2 - Effect Smart Market.
• Phase 3 - Effect Power.

Source

This Post is focused on how the phase one of the Cryptoproject development, Effect Mechanical Turk (EMT). For a more general overview of this blockchain project, click here or check out one of the information channels listed below.

A Brief History of The Mechanical Turk.

The Idea of creating an on demand online workforce was birthed by Venky Harinarayan, a Computer Science PhD holder of Standford University and a Bachelor in Technology of Computer Science from IIT Madras. MTurk was launched 2nd November of the year 2005. The user base grew rapidly too recording over a 100,000 workers in over a 100 countries. Amazon's Mechanical Turk (AMT) is one of the well known mechanical turks around now.

What is a Mechanical Turk?

This online crowdsourced on demand workforce concept derives it's name from "The Turk", an 18th century delusional humanoid chess-playing automaton created by Wolfgang von Kempelen which toured Europe defeating notable chess champions. Delusional because von Kempelen deceived Chess competitors that it had a mind of it's own. It was built to secretly house a human-aid (usually a chess champion).
The notion of a human aid is still what is under consideration, although not necessarily a chess playing champion.😏

A Mechanical Turk is an Online platform where requesters can create microtasks for workers also known as Turkers. It creates a large global workforce for the requesters which can be individuals or Corporations. Created to source for Human Intelligence (HI) through Human Intelligence Tasks (HITs) on a global level, It has been applied in a lot of ways which include Social experiments, Artistic or Educational Reseraches, Missing Persons searches, Information gathering, Audio transcribing e.t.c.

But how does this fit into the effect.ai project plan?

Big Clean Data is usually the key behind a well trained machine in a particular Domain. These Data don't just fall out of the sky to be used, they are usually gathered and cleaned using HI before fed into the algorithms that train the machines. Outliers are data points that fall out of the major cluster of data points when data is plotted on a scatter diagram. As few as they come, they are responsible for inconsistencies in the machine's predictive analysis. These big data are either gathered through turkers all over the globe through a platform like the EMT or are cleaned through it, in the form of image tags classification, podcasts transcription and translations, Data labelling e.t.c. Hence the Effect force provide HI aid for tasks that are not yet well suited for AI as it builds the user base of the network.

How The EMT Works.

source

The Effect requesters which may be an AI enthusiasts or large corporation, loads its datasets into the platform and create tasks on them which maybe a Data labelling tasks, Image classification or podcasts transcribing. Without any intermediary costs, the requester spends previously acquired EFX tokens from the Effect Galaxy Pool (EGP) on Smart Contracts (SCs) where he/she specifies the amount of EFX tokens assigned to each completed tasks. According the the whitepaper, the network uses a similar hash-based distributed file storage where each media asset uploaded by the requesters can be called up by a single hash or normal traditional storage channels such as Amazon S3, Google cloud storage or File Transfer Protocol (FTP) will also be available for requesters to supply and store Media datasets.

The EMT provides a feature which allows requesters to encrypt uploaded datasets inorder to allow only selected workers to view the uploaded datasets.

Registered workers are invoke any job contract of their choice and are only limited by the Effect Reputation Score (ERS). The ERS increases as they provide more quality completed tasks. A higher ERS allows workers to invoke more rewarding task contracts. The completed tasks are reviewed by the Effect Council (EC) which comprises of individuals voted for by majority to avoid unfair rejection of completed tasks by requesters. Requesters are also scored to increase their trustworthiness guarantee. A higher ERS for requesters allow for lower transactions fees. The ERS which expires overtime prompts users (Workers and Turkers) to remain active producing quality work.

The execution fees and SCs variables are governed by the EC to make sure it is up to date with the world's dynamic market. Improvements are also implemented when voted in by majority of the EC before the expiration time. From these above features it can be seen that the project is geared up towards solving the problems of inequivalent compensation for completed tasks, reward fraud by the requesters e.t.c associated with MTurks in a dynamic world.
In addition to these features, the Effect network is hosted on the NEO blockchain which uses Delegate Byzantine Fault Tolerance (dBFT) consensus and allows for over 10,000 transactions per second, user identification, file storage and Turing-Complete Smart Contracts.

Source

The EMT Effects.

The growth AI industry which is envisioned to contribute about $15.7 trillion to the world economy by 2030 might come to a halt if big clean data continue to remain behind closed doors of big corporations. The EMT rules out this possibility as it creates a fairly priced cost, scalable and secure platform for the creation of new big clean data. As the platform aids Datasets creation, it also increases the entry of AI enthusiasts into the AI industry by elevating them above their environments devoid of data and packed up with AI critics into an AI data creating and processing workforce space with the capability of handling micro-tasks regardless of it's divergence and by cutting data cleaning costs.
Still on EMT effect on the AI industry, the platform enhances collaboration between AI and HI. It allows data verification, filtering and structuring which are tasks suited for HI to be completed by Turkers which are then used as training data for the AI which are well suited for analysis and correlations.

EMT increases the reputation blockchain technology. Successfully implementing these microtasks which will in no doubt raise the standard of living by providing at least an hourly income of $1 would attract attention from a whole lot of industries more than it has now. These industries would seek new ways of breaking truckload of tasks into microtasks which are to be completed by turkers. This would increase the embrace of this next generation technology and pique the interest of more beneficiaries and benefactors. It would not suprise me if in the nearest future, Blockchain101 is included in the curriculum of most schools.

Clearly, this is an innovative and feasible blockchain solution to reducing the crime wave in underdeveloped, developing and even developed countries . The connection between between unemployment rates and crime waves is undisputable. This is not just because an idle mind is the devil's workshop but also because the number of desperate and bribable increases along with the increased unemployment rates. Crime waves is only the stem of this multi-headed unemployment hydra. Some consequences of increased crime waves are brain drain, a devoid of investors economy, raised mortality rates, devalued currency e.t.c.

Therefore the employments provided by this first phase of the Effect Network in the form of microtasks will serve as a very effective weapon to slay this multi-headed hydra by providing optimal income from the EGP to her turkers, reducing the cost of completing an AI project and hence the cost of obtaining these completed AI services. Therefore we could say that it reduces crime waves by providing employment and funds and also by helping to save funds.

In Summary,

The first phase of the Effect Network sets in motion a chain of events which aids a mutual interaction between the AI industry and her prospective user base and their environments. The implications of these events are so far reaching that one can not really predict its end. However one thing is sure, a determined collaborative effort always yields result. Therefore it is definite that the joint efforts of both HI and AI will lead to definite progress.

robo-advisor(1).jpg
Source

Further Information and Resources Channels.

effect2018.