A blockchain based AI competition sets a new use case for the data science industry

Companies in need of a data scientist can publish their problem on a blockchain to get it solved in perspective
27 February 2018   1325

Algorithmia today launched a very interesting competition on Ethereum platform - an AI competition. The task is to design the best predictor for voters preference in the last US presidential election classified by longitude and latitude. The bounty for the creator of such a model is 3 ETH (~ $2,500 at the time of publication). The task has a form of a smart contract of a special type called DanKu - Algorithmia's own creation.

DanKu allows a task creator to code a smart contract with a problem that deals with machine learning, publish a hidden dataset made available once the model creator wishes to evaluate the performance of the designed model. Then, through the smart contract, the task creator establishes randomized training sets made publicly available. All of the computation to test the model are performed on Ethereum blockchain. Because it is a smart contract, both parties don't need to trust each other: once the conditions of the problem are satisfied, the task creator gets the designed model, and the model creator gets his reward.

Apart from the novelty of such use case, it is also a proof-of-concept of a unique service useful for the industry. Say, a company needs a machine learning solution to its problem, but doesn't have the resources to hire a data scientist. Then, with Algorithmia and its DanKu smart contract that company can create a smart contract and request a specific machine learning approach to solve the problem. This is the first time neural network inference is run on top of Ethereum.

If a machine knew how to define a contract automatically, and put it on some system that other systems could solve, there is a potential for machine learning systems solving problems autonomously at a certain level.


Diego Oppenheimer

Co-founder and CEO, Algorithmia

We Need 100k Transactions\Sec, - Buterin

During the conversation with the head of Abra Vitalik noted that the developers are focused on several main issues - scalability, privacy and ease of use
21 March 2019   104

The founder of Ethereum, Vitaly Buterin, expressed the opinion that for the network of the second most capitalized cryptocurrency, it is vital in the long run to achieve a throughput of 100 thousand transactions per second. It is reported by The Daily Hodl.

During the conversation with the head of Abra, Bill Barkhidt, Vitalik noted that the developers are focused on several main issues - scalability, privacy and ease of use.

As far as the big problems, my top three at this point are probably scalability, privacy and usability. So scalability – the Ethereum blockchain right now can process 15 transactions per second. Really, we need 100,000.

Vitalik Buterin

Creator, Ethereum 

According to him, in the context of solving the problem of scaling, the Ethereum team has high hopes for sharding.

There are two major kinds of strategies that we’re working on for scalability. One is layer-one scaling and the other is layer-two scaling. Layer-one scaling basically means improving the blockchain protocol itself to process a larger set of transactions. And the main bottleneck with blockchains right now is basically every user has to download the whole blockchain. Which basically means the blockchain can’t hold more transactions than one guy’s computer can store.

And our solution to this, called sharding, basically means that you split up the different transactions to randomly selected, different groups of computers. And this basically means that the blockchain can process way more things than one single computer can hold. And that can increase scalability by maybe a factor of 1,000 or so, but then potentially even more, much later down the road

Vitalik Buterin

Creator, Ethereum 

According to Vitalik, Ethereum developers also continue active work on second-level scaling solutions.

So there’s two major classes of systems we’re working on in this regard. One is called state channels. And there’s a bunch of teams working on this. There’s a team called L4 in Toronto that’s done some really good work. And another project is Plasma. And there’s a lot of work that’s been done on that. OmiseGo is this decentralized exchange that’s building on plasma. There’s TheMatter. There’s more and more of these projects. And then, there’s one of our researchers, Karl Floersch, who has been working on implementation of a reasonably complete Plasma prime specification, which is the latest version of Plasma – which has some really cool features in terms of increasing scalability and reducing the amount of data you have to store.

Vitalik Buterin

Creator, Ethereum 

Discussing the problem of scaling, Buterin mentioned Bitcoin. According to him, developers should achieve a significant increase bandwidth of Ethereum network. Otherwise, the second cryptocurrency capitalization will not be actively involved in everyday transactions and a significant market share will remain behind BTC.

If Bitcoin wishes to just be a store of value, then realistically it’s probably fine, though I think they should switch to proof-of-stake. If they want to actually be a currency that people use for transactions, then I do think base-layer scaling, and also speeding up the blockchain and reducing block times at the base layer, is also something which is very important.

Vitalik Buterin

Creator, Ethereum 

On the last day of winter hardfork Constantinople hardfork took place on the Ethereum network.